# Step 5 Preview

> Source: Benchize (https://benchize.com) — https://benchize.com/models/step-5-preview. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/step-5-preview.

StepFun

Canonical URL: https://benchize.com/models/step-5-preview
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

StepFun

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [Arena text style control / multi turn](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Amulti_turn.md) | 1,443.89 | — | 64 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1443.8863986264273 \| rating \| reported \| Higher is better \| ### Configuration category multi_turn config text_style_control ### Reported statistics lower 1426.8673773094788 rank 101 upper 1460.9054199433758 variance 75.40028413851745 votes 1231 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9416/row` |  |  |  |  |  |  |  |
| [Arena vision / homework (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Ahomework--config-vision.md) | 1,265.3 | — | 52 / 73 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1265.2969229847035 \| rating \| reported \| Higher is better \| ### Configuration category homework config vision ### Reported statistics lower 1227.5067974116225 rank 66 upper 1303.0870485577846 variance 371.7581412394755 votes 233 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/873/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__international_trade_sanctions.md) | 90.87% | SE ±1.82 pp | 9 / 49 | $0.9 | 1,780.01source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.873 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.89711 \| USD \| per_test \| Lower is better \| \| `latency` \| 1780.01 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.82 pp sourcePopulation 81 stderr 1.818 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__international_trade_sanctions/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text style control / industry business and management and financial operations](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_business_and_management_and_financial_operations.md) | 1,441.86 | — | 62 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1441.8587401505142 \| rating \| reported \| Higher is better \| ### Configuration category industry_business_and_management_and_financial_operations config text_style_control ### Reported statistics lower 1426.4002214377947 rank 99 upper 1457.3172588632337 variance 62.20704475710576 votes 1506 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/4381/row` |  |  |  |  |  |  |  |
| [Arena text / chinese](https://benchize.com/benchmarks/lmarena%3Atext%3Achinese.md) | 1,499.79 | — | 39 / 146 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1499.7891313088298 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text ### Reported statistics lower 1478.0850484470095 rank 53 upper 1521.4932141706502 variance 122.62716714158128 votes 710 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/469/row` |  |  |  |  |  |  |  |
| [ProofBench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Aproof_bench%3Aoverall.md) | 42% | SE ±4.96 pp | 26 / 34 | $0.71 | 2,313.59source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_proof_bench.i3Hwh8rW.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 42 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.714564 \| USD \| per_test \| Lower is better \| \| `latency` \| 2313.592 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.96 pp sourcePopulation 52 stderr 4.96 tokenTotals {"cache_read_tokens":234817792,"cache_write_tokens":null,"input_tokens":258573571,"output_tokens":13318410,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_proof_bench.i3Hwh8rW.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:46:01.401Z Evidence SHA-256 `22be2aca25de5c4a9eb6448a95713e09a7c97126dfc6f4f75c4afd0e563209cb` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / highlight js](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Ahighlight_js.md) | 30.77% | SE ±7.49 pp | 13 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 30.769 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.49 pp sourcePopulation 20 stderr 7.487 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/highlight_js/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [IOI / ioi2026](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aioi2026.md) | 37.67% | SE ±18.13 pp | 24 / 29 | $1.65 | 4,001.48source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 37.667 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.648748 \| USD \| per_test \| Lower is better \| \| `latency` \| 4001.475 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±18.13 pp sourcePopulation 44 stderr 18.13 tokenTotals {"cache_read_tokens":68157696,"cache_write_tokens":0,"input_tokens":70080870,"output_tokens":1326129,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/ioi2026/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena vision / ocr (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Aocr--config-vision.md) | 1,282.06 | — | 35 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1282.0568798452775 \| rating \| reported \| Higher is better \| ### Configuration category ocr config vision ### Reported statistics lower 1264.2111254359086 rank 42 upper 1299.9026342546465 variance 82.9036481463481 votes 1102 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/1060/row` |  |  |  |  |  |  |  |
| [Arena text / overall](https://benchize.com/benchmarks/lmarena%3Atext.md) | 1,440.32 | — | 48 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1440.3232927002698 \| rating \| reported \| Higher is better \| ### Configuration category overall config text ### Reported statistics lower 1433.4324098530083 rank 74 upper 1447.2141755475316 variance 12.3609984204139 votes 8756 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/73/row` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / general quantitative analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Ageneral_quantitative_analysis.md) | 74.81% | SE ±5.34 pp | 15 / 18 | $0.47 | 378.59source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.805 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.467778 \| USD \| per_test \| Lower is better \| \| `latency` \| 378.592 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.34 pp sourcePopulation 28 stderr 5.34 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/general_quantitative_analysis/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / bankruptcy restructuring](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Abankruptcy_restructuring.md) | 0% | SE ±0 pp | 3 / 49 | $1.02 | 1,782.29source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.021346 \| USD \| per_test \| Lower is better \| \| `latency` \| 1782.292 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/bankruptcy_restructuring/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SRE Bench / overall](https://benchize.com/benchmarks/vals-ai%3Asrebench%3Aoverall.md) | 0.38% | SE ±0.38 pp | 21 / 24 | $3.07 | 5,625.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0.382 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.06713 \| USD \| per_test \| Lower is better \| \| `latency` \| 5625.323 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.38 pp sourcePopulation 35 stderr 0.382 tokenTotals {"cache_read_tokens":9309212928,"cache_write_tokens":null,"input_tokens":9457799020,"output_tokens":70200517,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_sre_bench.Bd0uBJxH.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:15.724Z Evidence SHA-256 `a10b2c964f5878b0fc21c888703f311a5ef7d7a74fc3c529cab3ddd9fa474e2b` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / precedents](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aprecedents.md) | 31.08% | SE ±6.7 pp | 15 / 18 | $0.91 | 782.37source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 31.079 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.905855 \| USD \| per_test \| Lower is better \| \| `latency` \| 782.372 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.7 pp sourcePopulation 28 stderr 6.701 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/precedents/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Legal Research Bench / civil litigation](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Acivil_litigation.md) | 48.15% | SE ±9.8 pp | 5 / 50 | $0.92 | 2,136.96source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 48.148 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.917209 \| USD \| per_test \| Lower is better \| \| `latency` \| 2136.958 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.8 pp sourcePopulation 82 stderr 9.799 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/civil_litigation/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [IOI / ioi2025](https://benchize.com/benchmarks/vals-ai%3Aioi%3Aioi2025.md) | 49.83% | SE ±17.72 pp | 16 / 29 | $2.29 | 5,117.86source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 49.833 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.286536 \| USD \| per_test \| Lower is better \| \| `latency` \| 5117.861 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±17.72 pp sourcePopulation 44 stderr 17.718 tokenTotals {"cache_read_tokens":106455808,"cache_write_tokens":0,"input_tokens":109078778,"output_tokens":1669504,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_ioi.CuUolFNv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:46:11.757Z Evidence SHA-256 `e1545b21b5b5c98d037aefa9d441b65859d07a10550f242d0e16162a49b8e8ab` Source locator `/tasks/ioi2025/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Legal Research Bench / criminal](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Acriminal.md) | 38.6% | SE ±6.51 pp | 12 / 50 | $0.87 | 2,210.11source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 38.596 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.87429 \| USD \| per_test \| Lower is better \| \| `latency` \| 2210.107 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.51 pp sourcePopulation 82 stderr 6.505 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/criminal/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / energy natural resources](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aenergy_natural_resources.md) | 40% | SE ±24.5 pp | 1 / 49 | $1.16 | 1,767.97source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 40 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.160357 \| USD \| per_test \| Lower is better \| \| `latency` \| 1767.97 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±24.5 pp sourcePopulation 81 stderr 24.495 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/energy_natural_resources/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / grommet](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Agrommet.md) | 25% | SE ±9.93 pp | 10 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 25 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.93 pp sourcePopulation 20 stderr 9.934 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/grommet/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / instruction following](https://benchize.com/benchmarks/lmarena%3Atext%3Ainstruction_following.md) | 1,438.4 | — | 45 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1438.4031158485825 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text ### Reported statistics lower 1427.460113718719 rank 61 upper 1449.3461179784458 variance 31.17286978819115 votes 3099 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7594/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / family](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Afamily.md) | 9.09% | SE ±6.27 pp | 34 / 50 | $0.89 | 2,391.97source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 9.091 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.88974 \| USD \| per_test \| Lower is better \| \| `latency` \| 2391.966 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.27 pp sourcePopulation 82 stderr 6.273 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/family/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Tax Agent Bench / rule source lookup](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Arule_source_lookup.md) | 41.67% | SE ±8.47 pp | 15 / 46 | $0.55 | 1,297.97source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 41.667 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.552303 \| USD \| per_test \| Lower is better \| \| `latency` \| 1297.966 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.47 pp sourcePopulation 73 stderr 8.47 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/rule_source_lookup/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Code Migration v1.1 / cli](https://benchize.com/benchmarks/vals-ai%3Acode-migration-v1-1%3Acli.md) | 35.17% | SE ±2.5 pp | 13 / 17 | $10.46 | 13,290.26source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_code_migration_v1_1.bJHk6HEY.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 35.171 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 10.456511 \| USD \| per_test \| Lower is better \| \| `latency` \| 13290.264 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.5 pp sourcePopulation 26 stderr 2.503 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration_v1_1.bJHk6HEY.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:46:09.631Z Evidence SHA-256 `352faac603c73f05cc01e89214fb4adcafa05ea7432ea71ea9903a6b5fe36b08` Source locator `/tasks/cli/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena vision style control / overall](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Aoverall.md) | 1,247.18 | — | 44 / 82 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1247.1751302786683 \| rating \| reported \| Higher is better \| ### Configuration category overall config vision_style_control ### Reported statistics lower 1232.0164871773625 rank 58 upper 1262.3337733799742 variance 59.8169787568475 votes 1649 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/57/row` |  |  |  |  |  |  |  |
| [Arena text style control / french](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Afrench.md) | 1,444.16 | — | 72 / 132 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1444.1582917618725 \| rating \| reported \| Higher is better \| ### Configuration category french config text_style_control ### Reported statistics lower 1412.7655041575445 rank 118 upper 1475.5510793662002 variance 256.54501572721887 votes 388 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/2955/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass business commercial](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__business_commercial.md) | 89.08% | SE ±2.24 pp | 9 / 50 | $0.81 | 2,344.54source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.077 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.811263 \| USD \| per_test \| Lower is better \| \| `latency` \| 2344.543 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.24 pp sourcePopulation 82 stderr 2.243 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__business_commercial/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Code Migration / cli](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Acli.md) | 31.97% | SE ±2.63 pp | 25 / 47 | $9.17 | 14,495.57source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 31.969 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 9.165067 \| USD \| per_test \| Lower is better \| \| `latency` \| 14495.57 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.63 pp sourcePopulation 79 stderr 2.628 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/cli/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / comparables](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Acomparables.md) | 53.62% | SE ±6.48 pp | 14 / 18 | $0.8 | 683.52source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 53.62 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.804444 \| USD \| per_test \| Lower is better \| \| `latency` \| 683.515 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.48 pp sourcePopulation 28 stderr 6.479 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/comparables/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / funds asset management](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Afunds_asset_management.md) | 0% | SE ±0 pp | 11 / 49 | $1.26 | 2,127.48source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.258092 \| USD \| per_test \| Lower is better \| \| `latency` \| 2127.48 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/funds_asset_management/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass numbers calculations](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__numbers_calculations.md) | 72.08% | SE ±6.94 pp | 29 / 46 | $0.24 | 647.94source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.083 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.235703 \| USD \| per_test \| Lower is better \| \| `latency` \| 647.942 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.94 pp sourcePopulation 73 stderr 6.942 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__numbers_calculations/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / financial modeling](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Afinancial_modeling.md) | 29.33% | SE ±6.18 pp | 14 / 18 | $0.53 | 508.54source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 29.33 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.53266 \| USD \| per_test \| Lower is better \| \| `latency` \| 508.535 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.18 pp sourcePopulation 28 stderr 6.184 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/financial_modeling/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / earnings analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aearnings_analysis.md) | 85.13% | SE ±4.72 pp | 8 / 18 | $0.22 | 376.22source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 85.131 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.219532 \| USD \| per_test \| Lower is better \| \| `latency` \| 376.22 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.72 pp sourcePopulation 28 stderr 4.722 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/earnings_analysis/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB / scratch](https://benchize.com/benchmarks/vals-ai%3Aemb%3Ascratch.md) | 66.78% | SE ±2.48 pp | 18 / 47 | $1.22 | 2,506.87source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 66.777 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.217876 \| USD \| per_test \| Lower is better \| \| `latency` \| 2506.874 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.48 pp sourcePopulation 76 stderr 2.483 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/scratch/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB / template](https://benchize.com/benchmarks/vals-ai%3Aemb%3Atemplate.md) | 47.42% | SE ±5.43 pp | 35 / 47 | $1.44 | 2,723.32source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.417 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.44076 \| USD \| per_test \| Lower is better \| \| `latency` \| 2723.319 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.43 pp sourcePopulation 76 stderr 5.427 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/template/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text style control / spanish](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aspanish.md) | 1,432.96 | — | 59 / 134 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1432.9592682587127 \| rating \| reported \| Higher is better \| ### Configuration category spanish config text_style_control ### Reported statistics lower 1400.6322769806443 rank 99 upper 1465.286259536781 variance 272.0410171944726 votes 318 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10863/row` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass.md) | 59.54% | SE ±3.09 pp | 27 / 46 | $0.55 | 1,288.52source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.539 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.55338 \| USD \| per_test \| Lower is better \| \| `latency` \| 1288.515 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.09 pp sourcePopulation 73 stderr 3.09 tokenTotals {"cache_read_tokens":262109696,"cache_write_tokens":null,"input_tokens":311523716,"output_tokens":16401077,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / industry mathematical](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_mathematical.md) | 1,459.67 | — | 48 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1459.6729869024084 \| rating \| reported \| Higher is better \| ### Configuration category industry_mathematical config text ### Reported statistics lower 1434.25369900175 rank 67 upper 1485.0922748030669 variance 168.2017763397004 votes 502 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/5979/row` |  |  |  |  |  |  |  |
| [Arena text style control / instruction following](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ainstruction_following.md) | 1,443.12 | — | 49 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1443.1237317322682 \| rating \| reported \| Higher is better \| ### Configuration category instruction_following config text_style_control ### Reported statistics lower 1432.1503278418536 rank 76 upper 1454.0971356226828 variance 31.346318823849106 votes 3099 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/7609/row` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / physical sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aphysical_sciences.md) | 11.77% | SE ±8.06 pp | 13 / 30 | — | — | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 11.765 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.06 pp sourcePopulation 41 stderr 8.055 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label stepfun/step-5-preview Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/physical_sciences/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / overall](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aoverall.md) | 50.67% | SE ±0.32 pp | 31 / 50 | $0.62 | 747.59source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 50.671 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.619283 \| USD \| per_test \| Lower is better \| \| `latency` \| 747.585 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.32 pp sourcePopulation 80 stderr 0.324 tokenTotals {"cache_read_tokens":254565035,"cache_write_tokens":null,"input_tokens":464025197,"output_tokens":20770861,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate healthcare life sciences](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__healthcare_life_sciences.md) | 90.1% | SE ±1.72 pp | 16 / 49 | $1.29 | 2,200.13source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 90.099 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.289815 \| USD \| per_test \| Lower is better \| \| `latency` \| 2200.132 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.72 pp sourcePopulation 81 stderr 1.719 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__healthcare_life_sciences/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / lighthouse](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Alighthouse.md) | 24.07% | SE ±5.87 pp | 7 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 24.074 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.87 pp sourcePopulation 20 stderr 5.873 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/lighthouse/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / industry software and it services](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_software_and_it_services.md) | 1,473.14 | — | 41 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1473.1412313872343 \| rating \| reported \| Higher is better \| ### Configuration category industry_software_and_it_services config text ### Reported statistics lower 1462.5298386667455 rank 55 upper 1483.7526241077235 variance 29.312212033058003 votes 3192 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6755/row` |  |  |  |  |  |  |  |
| [Arena text / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_medicine_and_healthcare.md) | 1,438.7 | — | 59 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1438.7024251892426 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text ### Reported statistics lower 1414.817933666583 rank 100 upper 1462.5869167119026 variance 148.50320201868868 votes 614 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/6412/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass administrative regulatory](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__administrative_regulatory.md) | 87.76% | SE ±2.3 pp | 15 / 50 | $0.86 | 2,171.35source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.76 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.862352 \| USD \| per_test \| Lower is better \| \| `latency` \| 2171.348 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.3 pp sourcePopulation 82 stderr 2.3 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__administrative_regulatory/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / russian](https://benchize.com/benchmarks/lmarena%3Atext%3Arussian.md) | 1,432.87 | — | 50 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1432.865944693527 \| rating \| reported \| Higher is better \| ### Configuration category russian config text ### Reported statistics lower 1416.7938137337323 rank 77 upper 1448.9380756533212 variance 67.24356700044173 votes 1288 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/10460/row` |  |  |  |  |  |  |  |
| [Arena text / longer query](https://benchize.com/benchmarks/lmarena%3Atext%3Alonger_query.md) | 1,448.58 | — | 46 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1448.5836104482355 \| rating \| reported \| Higher is better \| ### Configuration category longer_query config text ### Reported statistics lower 1438.381055531172 rank 64 upper 1458.7861653652992 variance 27.09703050178487 votes 3705 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/8581/row` |  |  |  |  |  |  |  |
| [Arena text style control / russian](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Arussian.md) | 1,436.79 | — | 58 / 150 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1436.7858943414108 \| rating \| reported \| Higher is better \| ### Configuration category russian config text_style_control ### Reported statistics lower 1420.7081229483383 rank 96 upper 1452.8636657344832 variance 67.29077286352111 votes 1288 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/10479/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate capital markets](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__capital_markets.md) | 92.16% | SE ±1.69 pp | 9 / 49 | $0.92 | 1,604.09source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 92.157 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.917971 \| USD \| per_test \| Lower is better \| \| `latency` \| 1604.092 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.69 pp sourcePopulation 81 stderr 1.687 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__capital_markets/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Code Migration / code quality](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Acode_quality.md) | 4.38% | SE ±0.04 pp | 13 / 47 | $7.61 | 12,856.9source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 4.379 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 7.606885 \| USD \| per_test \| Lower is better \| \| `latency` \| 12856.895 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.04 pp sourcePopulation 79 stderr 0.036 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/code_quality/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB / ma](https://benchize.com/benchmarks/vals-ai%3Aemb%3Ama.md) | 64.61% | SE ±8.54 pp | 25 / 47 | $1.11 | 2,412.2source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.61 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.114729 \| USD \| per_test \| Lower is better \| \| `latency` \| 2412.197 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.54 pp sourcePopulation 76 stderr 8.543 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/ma/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / hard prompts](https://benchize.com/benchmarks/lmarena%3Atext%3Ahard_prompts.md) | 1,461.5 | — | 40 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1461.4965567653248 \| rating \| reported \| Higher is better \| ### Configuration category hard_prompts config text ### Reported statistics lower 1453.116073030746 rank 55 upper 1469.8770404999036 variance 18.282769880857423 votes 5502 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/3505/row` |  |  |  |  |  |  |  |
| [Arena text / expert](https://benchize.com/benchmarks/lmarena%3Atext%3Aexpert.md) | 1,484.53 | — | 35 / 148 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1484.527349608474 \| rating \| reported \| Higher is better \| ### Configuration category expert config text ### Reported statistics lower 1465.0250723211611 rank 45 upper 1504.0296268957866 variance 99.00895392717263 votes 874 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/2515/row` |  |  |  |  |  |  |  |
| [Legal Research Bench / health](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Ahealth.md) | 53.33% | SE ±9.26 pp | 24 / 50 | $0.91 | 2,028.96source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 53.333 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.905473 \| USD \| per_test \| Lower is better \| \| `latency` \| 2028.962 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.26 pp sourcePopulation 82 stderr 9.264 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/health/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate corporate governance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__corporate_governance.md) | 89.38% | SE ±1.81 pp | 13 / 49 | $0.87 | 1,546.64source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.384 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.874361 \| USD \| per_test \| Lower is better \| \| `latency` \| 1546.64 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.81 pp sourcePopulation 81 stderr 1.806 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__corporate_governance/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / capital markets](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acapital_markets.md) | 20% | SE ±20 pp | 4 / 49 | $0.92 | 1,604.09source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.917971 \| USD \| per_test \| Lower is better \| \| `latency` \| 1604.092 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±20 pp sourcePopulation 81 stderr 20 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/capital_markets/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate banking finance](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__banking_finance.md) | 89.59% | SE ±1.87 pp | 17 / 49 | $0.83 | 1,611.05source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.591 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.82827 \| USD \| per_test \| Lower is better \| \| `latency` \| 1611.048 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.87 pp sourcePopulation 81 stderr 1.865 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__banking_finance/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / all pass](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aall_pass.md) | 44.67% | SE ±2.16 pp | 15 / 18 | $0.58 | 542.51source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 44.665 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.575115 \| USD \| per_test \| Lower is better \| \| `latency` \| 542.506 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 28 stderr 2.162 tokenTotals {"cache_read_tokens":214876672,"cache_write_tokens":null,"input_tokens":407803894,"output_tokens":20205838,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/all_pass/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text / industry writing and literature and language](https://benchize.com/benchmarks/lmarena%3Atext%3Aindustry_writing_and_literature_and_language.md) | 1,424.81 | — | 48 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1424.805589491675 \| rating \| reported \| Higher is better \| ### Configuration category industry_writing_and_literature_and_language config text ### Reported statistics lower 1412.1939185012282 rank 71 upper 1437.4172604821215 variance 41.40464667080087 votes 2398 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/7188/row` |  |  |  |  |  |  |  |
| [MysteryMechanism / overall](https://benchize.com/benchmarks/vals-ai%3Amysterymechanism%3Aoverall.md) | 18.47% | SE ±2.61 pp | 17 / 19 | $0.6 | 1,760.42source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_mysterymechanism.75uyyp4P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 18.468 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.597179 \| USD \| per_test \| Lower is better \| \| `latency` \| 1760.418 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.61 pp sourcePopulation 27 stderr 2.61 tokenTotals {"cache_read_tokens":419705856,"cache_write_tokens":0,"input_tokens":455746580,"output_tokens":27980610,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_mysterymechanism.75uyyp4P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:04.766Z Evidence SHA-256 `1ef09cd0a0ee8690b0c07b2272057461242359b1680e24975ea42971e03acea9` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text style control / industry medicine and healthcare](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Aindustry_medicine_and_healthcare.md) | 1,453.51 | — | 65 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1453.5094065379064 \| rating \| reported \| Higher is better \| ### Configuration category industry_medicine_and_healthcare config text_style_control ### Reported statistics lower 1429.435004737184 rank 105 upper 1477.5838083386288 variance 150.8741468060038 votes 614 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/6417/row` |  |  |  |  |  |  |  |
| [Arena text / creative writing](https://benchize.com/benchmarks/lmarena%3Atext%3Acreative_writing.md) | 1,408.57 | — | 49 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1408.5710269682559 \| rating \| reported \| Higher is better \| ### Configuration category creative_writing config text ### Reported statistics lower 1394.190377712197 rank 82 upper 1422.9516762243147 variance 53.83451513568913 votes 1871 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `8e027ef69b8de06c4b2a8b217bfe816b1f1536269bb257fd295b4ef1ba8246ac` Source locator `/rows/1303/row` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate white collar defense investigations](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__white_collar_defense_investigations.md) | 86.53% | SE ±1.98 pp | 13 / 49 | $0.84 | 1,482.31source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 86.532 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.835096 \| USD \| per_test \| Lower is better \| \| `latency` \| 1482.306 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.98 pp sourcePopulation 81 stderr 1.984 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__white_collar_defense_investigations/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Vals Science Index / biomysterybench](https://benchize.com/benchmarks/vals-ai%3Ascience_index%3Abiomysterybench.md) | 67.41% | SE ±0.37 pp | 12 / 17 | $0.49 | 1,294.75source duration | [Vals AI · 1.0](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.407 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.494625 \| USD \| per_test \| Lower is better \| \| `latency` \| 1294.751 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.37 pp sourcePopulation 21 stderr 0.37 tokenTotals {"cache_read_tokens":257300480,"cache_write_tokens":0,"input_tokens":270017689,"output_tokens":7012589,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_science_index.CN7FN-f1.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.0 Observed 2026-10-21T16:45:50.879Z Evidence SHA-256 `fa4313008e91752fcfc185aaa78665f780d0fc23d7c7198a9853dbf2eaaaef1f` Source locator `/tasks/biomysterybench/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / market analysis](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Amarket_analysis.md) | 65.17% | SE ±6.57 pp | 7 / 18 | $0.3 | 361.65source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.172 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.302516 \| USD \| per_test \| Lower is better \| \| `latency` \| 361.647 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.57 pp sourcePopulation 28 stderr 6.571 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/market_analysis/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB v1.1 / template](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Atemplate.md) | 72.22% | SE ±2.99 pp | 11 / 19 | $1.44 | 2,723.32source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 72.224 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.44076 \| USD \| per_test \| Lower is better \| \| `latency` \| 2723.319 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.99 pp sourcePopulation 30 stderr 2.986 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/template/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Tax Agent Bench / weighted pass fact pattern analysis](https://benchize.com/benchmarks/vals-ai%3Atax_agent_bench%3Aweighted_pass__fact_pattern_analysis.md) | 56.37% | SE ±7.17 pp | 28 / 46 | $0.53 | 1,302.26source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 56.368 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.530547 \| USD \| per_test \| Lower is better \| \| `latency` \| 1302.263 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±7.17 pp sourcePopulation 73 stderr 7.173 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_tax_agent_bench.BFpexnpS.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:55.958Z Evidence SHA-256 `446db182b67efb7923b3187a4040ec3a3ea492b8ad699a2b6d708d714cc063ac` Source locator `/tasks/weighted_pass__fact_pattern_analysis/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB v1.1 / operating](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Aoperating.md) | 68.85% | SE ±4.45 pp | 12 / 19 | $1.05 | 2,211.96source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 68.853 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.049678 \| USD \| per_test \| Lower is better \| \| `latency` \| 2211.962 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.45 pp sourcePopulation 30 stderr 4.445 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/operating/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text style control / coding](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Acoding.md) | 1,508.72 | — | 40 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1508.7244491033853 \| rating \| reported \| Higher is better \| ### Configuration category coding config text_style_control ### Reported statistics lower 1495.7038469627526 rank 61 upper 1521.7450512440178 variance 44.13325458321942 votes 2074 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/870/row` |  |  |  |  |  |  |  |
| [Arena text style control / japanese](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Ajapanese.md) | 1,388.2 | — | 59 / 120 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1388.2043921683132 \| rating \| reported \| Higher is better \| ### Configuration category japanese config text_style_control ### Reported statistics lower 1343.2796685720737 rank 101 upper 1433.129115764553 variance 525.3813419334924 votes 168 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/8051/row` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / precedents](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aprecedents.md) | 21.14% | SE ±0.79 pp | 35 / 50 | $1.08 | 1,404.17source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 21.142 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.077508 \| USD \| per_test \| Lower is better \| \| `latency` \| 1404.168 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.79 pp sourcePopulation 80 stderr 0.79 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/precedents/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / financial modeling](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Afinancial_modeling.md) | 16.71% | SE ±1.52 pp | 36 / 50 | $0.57 | 703.99source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 16.714 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.569585 \| USD \| per_test \| Lower is better \| \| `latency` \| 703.989 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.52 pp sourcePopulation 80 stderr 1.524 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/financial_modeling/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena vision style control / homework](https://benchize.com/benchmarks/lmarena%3Avision_style_control%3Ahomework.md) | 1,261.01 | — | 52 / 73 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1261.0124055978738 \| rating \| reported \| Higher is better \| ### Configuration category homework config vision_style_control ### Reported statistics lower 1222.9559299703553 rank 66 upper 1299.0688812253925 variance 377.0170148345404 votes 233 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `38568936995f36625aae49e9ea05557d55c1b1df9b2954f5c50cee523f26fe92` Source locator `/rows/873/row` |  |  |  |  |  |  |  |
| [Vibe Code Bench v1.1 / overall](https://benchize.com/benchmarks/vals-ai%3Avibe-code%3Aoverall.md) | 70.4% | SE ±4.27 pp | 31 / 59 | $5.86 | 5,466.36source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_vibe_code_bench.DcEBMbCA.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 70.399 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 5.856257 \| USD \| per_test \| Lower is better \| \| `latency` \| 5466.358 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness OpenHands maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±4.27 pp sourcePopulation 116 stderr 4.267 tokenTotals {"cache_read_tokens":2351680256,"cache_write_tokens":null,"input_tokens":2504815838,"output_tokens":8182685,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_vibe_code_bench.DcEBMbCA.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:46:08.625Z Evidence SHA-256 `0e8c23f521a0f751c5592952620e197b376cade41c78578461b7aa5740a173e7` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB v1.1 / ma](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Ama.md) | 74.65% | SE ±5.3 pp | 12 / 19 | $1.11 | 2,412.2source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 74.652 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.114729 \| USD \| per_test \| Lower is better \| \| `latency` \| 2412.197 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±5.3 pp sourcePopulation 30 stderr 5.295 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/ma/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Code Migration v1.1 / cobol](https://benchize.com/benchmarks/vals-ai%3Acode-migration-v1-1%3Acobol.md) | 82.5% | SE ±11.48 pp | 10 / 17 | $3.08 | 7,378.94source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_code_migration_v1_1.bJHk6HEY.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 82.5 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 3.082158 \| USD \| per_test \| Lower is better \| \| `latency` \| 7378.942 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±11.48 pp sourcePopulation 26 stderr 11.481 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration_v1_1.bJHk6HEY.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:46:09.631Z Evidence SHA-256 `352faac603c73f05cc01e89214fb4adcafa05ea7432ea71ea9903a6b5fe36b08` Source locator `/tasks/cobol/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / international trade sanctions](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Ainternational_trade_sanctions.md) | 0% | SE ±0 pp | 4 / 49 | $0.9 | 1,780.01source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.89711 \| USD \| per_test \| Lower is better \| \| `latency` \| 1780.01 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/international_trade_sanctions/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / intellectual property](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aintellectual_property.md) | 0% | SE ±0 pp | 6 / 49 | $0.74 | 1,316.85source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.743241 \| USD \| per_test \| Lower is better \| \| `latency` \| 1316.854 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/intellectual_property/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate real estate](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__real_estate.md) | 93% | SE ±1.38 pp | 10 / 49 | $1.1 | 1,656.61source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 93.003 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.096893 \| USD \| per_test \| Lower is better \| \| `latency` \| 1656.608 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.38 pp sourcePopulation 81 stderr 1.379 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__real_estate/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Code Migration / cobol](https://benchize.com/benchmarks/vals-ai%3Acode-migration%3Acobol.md) | 60% | SE ±16.33 pp | 29 / 47 | $2.93 | 7,940.87source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 60 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 2.932338 \| USD \| per_test \| Lower is better \| \| `latency` \| 7940.872 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±16.33 pp sourcePopulation 79 stderr 16.33 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_code_migration.Dttzoqk3.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.712Z Evidence SHA-256 `99194597f2c76c402195be47839409046e18bd30454616da6dd40b736a088fa4` Source locator `/tasks/cobol/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / carbon](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Acarbon.md) | 24.81% | SE ±3.76 pp | 8 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 24.812 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.76 pp sourcePopulation 20 stderr 3.759 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/carbon/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [SWE-bench Multimodal / scratch gui](https://benchize.com/benchmarks/vals-ai%3Aswebench-multimodal%3Ascratch_gui.md) | 16.67% | SE ±16.67 pp | 6 / 14 | — | — | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 16.667 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±16.67 pp sourcePopulation 20 stderr 16.667 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_swebench_multimodal.C7UOLrm7.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:12.211Z Evidence SHA-256 `5302c42065b017d7deaa43d9e34be1ec69893b7ec4cceb6c07bfbabbd20b1f82` Source locator `/tasks/scratch_gui/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / adjustments](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aadjustments.md) | 58.5% | SE ±6.89 pp | 13 / 18 | $0.59 | 563.34source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.495 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.586728 \| USD \| per_test \| Lower is better \| \| `latency` \| 563.336 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.89 pp sourcePopulation 28 stderr 6.891 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/adjustments/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Terminal-Bench-Science / earth sciences](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-science%3Aearth_sciences.md) | 0% | SE ±0 pp | 14 / 30 | — | — | [Vals AI · 0.1](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Mini-SWE-agent maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 41 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_science.o-HR8Nyt.json) Source Vals AI Original model label stepfun/step-5-preview Source version 0.1 Observed 2026-10-21T16:46:03.529Z Evidence SHA-256 `a30a8618deaecf018ea1c8c9f675576cdd9ab79742d05788067db49afcc5f91b` Source locator `/tasks/earth_sciences/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena vision / diagram (config: vision)](https://benchize.com/benchmarks/benchmark%3Avision%3Adiagram--config-vision.md) | 1,307.7 | — | 25 / 74 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1307.7041122784144 \| rating \| reported \| Higher is better \| ### Configuration category diagram config vision ### Reported statistics lower 1278.744517931741 rank 27 upper 1336.6637066250873 variance 218.31760897864393 votes 423 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/vision/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `06e61305486dce3fec5dc776ac2fd8321104e84c346cf9e08fa6f8e8a66ce0d7` Source locator `/rows/493/row` |  |  |  |  |  |  |  |
| [EMB v1.1 / scratch](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Ascratch.md) | 66.57% | SE ±2.53 pp | 11 / 19 | $1.22 | 2,506.87source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 66.568 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.217876 \| USD \| per_test \| Lower is better \| \| `latency` \| 2506.874 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.53 pp sourcePopulation 30 stderr 2.529 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/scratch/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / hardware](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Ahardware.md) | 20% | SE ±11.55 pp | 13 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 20 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±11.55 pp sourcePopulation 52 stderr 11.547 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label stepfun/step-5-preview Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/hardware/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Terminal-Bench 4.0 / science](https://benchize.com/benchmarks/vals-ai%3Aterminal-bench-4%3Ascience.md) | 33.33% | SE ±6.3 pp | 14 / 36 | — | — | [Vals AI · 4.0](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 33.333 \| percent \| reported \| Higher is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.3 pp sourcePopulation 52 stderr 6.299 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_terminal_bench_4.CAnGyzC_.json) Source Vals AI Original model label stepfun/step-5-preview Source version 4.0 Observed 2026-10-21T16:46:10.642Z Evidence SHA-256 `98f6964580e901f1aa143db664bd92aa558415f643569492545d67c47d7f519b` Source locator `/tasks/science/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / adjustments](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aadjustments.md) | 47.75% | SE ±2.66 pp | 31 / 50 | $0.6 | 1,058.39source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 47.752 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.603494 \| USD \| per_test \| Lower is better \| \| `latency` \| 1058.392 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.66 pp sourcePopulation 80 stderr 2.656 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/adjustments/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent (v2) / all pass](https://benchize.com/benchmarks/vals-ai%3Afabv2%3Aall_pass.md) | 38.81% | SE ±0.37 pp | 29 / 50 | $0.62 | 747.59source duration | [Vals AI · 2](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 38.809 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.619283 \| USD \| per_test \| Lower is better \| \| `latency` \| 747.585 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.37 pp sourcePopulation 80 stderr 0.368 tokenTotals {"cache_read_tokens":254565035,"cache_write_tokens":null,"input_tokens":464025197,"output_tokens":20770861,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2.DypCOnwd.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2 Observed 2026-10-21T16:45:57.035Z Evidence SHA-256 `f731b20a26c126725d125519f2b5841aa45254656dba859a9570d7ce7e5d8191` Source locator `/tasks/all_pass/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate tax](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__tax.md) | 87.29% | SE ±1.93 pp | 5 / 49 | $1.29 | 2,489.23source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 87.291 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.290764 \| USD \| per_test \| Lower is better \| \| `latency` \| 2489.228 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.93 pp sourcePopulation 81 stderr 1.929 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__tax/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / environmental esg](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Aenvironmental_esg.md) | 0% | SE ±0 pp | 1 / 49 | $0.98 | 1,922.72source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 0 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.97515 \| USD \| per_test \| Lower is better \| \| `latency` \| 1922.718 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0 pp sourcePopulation 81 stderr 0 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/environmental_esg/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [EMB v1.1 / dcf](https://benchize.com/benchmarks/vals-ai%3Aemb-v1-1%3Adcf.md) | 65.83% | SE ±6.9 pp | 7 / 19 | $1.21 | 2,762.16source duration | [Vals AI · 1.1](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.831 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.211646 \| USD \| per_test \| Lower is better \| \| `latency` \| 2762.159 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±6.9 pp sourcePopulation 30 stderr 6.896 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb_v1_1.11L06J7L.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1.1 Observed 2026-10-21T16:45:55.456Z Evidence SHA-256 `eea2ef2c83a6dc7d553c7c6531c287710ea7ffbc56ee1ccec06c85b104b2408e` Source locator `/tasks/dcf/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Harvey's Legal Agent Benchmark v1 / criteria pass rate arbitration international dispute resolution](https://benchize.com/benchmarks/vals-ai%3Ahlab-v1%3Acriteria_pass_rate__arbitration_international_dispute_resolution.md) | 89.25% | SE ±1.86 pp | 15 / 49 | $1.33 | 1,787.72source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 89.247 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.332675 \| USD \| per_test \| Lower is better \| \| `latency` \| 1787.716 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.86 pp sourcePopulation 81 stderr 1.858 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_agent_benchmark_v1.ZWRzZKcv.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:53.909Z Evidence SHA-256 `65e7cd51e83066a42c9d782635dba5f2e9e1b189bd8118d3510cd17d2ddcba3f` Source locator `/tasks/criteria_pass_rate__arbitration_international_dispute_resolution/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Legal Research Bench / weighted pass immigration](https://benchize.com/benchmarks/vals-ai%3Alegal_research%3Aweighted_pass__immigration.md) | 77.59% | SE ±8.49 pp | 31 / 50 | $1.13 | 2,795.76source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 77.592 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.125625 \| USD \| per_test \| Lower is better \| \| `latency` \| 2795.76 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±8.49 pp sourcePopulation 82 stderr 8.491 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_research.DiuL0R8P.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:52.389Z Evidence SHA-256 `26c58dfd47d0d2dc86f33c6d056f30309e0ee63570e7cd752c593b385f9d862e` Source locator `/tasks/weighted_pass__immigration/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [BioMysteryBench / overall](https://benchize.com/benchmarks/vals-ai%3Abiomysterybench%3Aoverall.md) | 67.41% | SE ±0.37 pp | 12 / 20 | $0.49 | 1,294.75source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.407 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.494625 \| USD \| per_test \| Lower is better \| \| `latency` \| 1294.751 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Terminus 2 maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.37 pp sourcePopulation 27 stderr 0.37 tokenTotals {"cache_read_tokens":257300480,"cache_write_tokens":0,"input_tokens":270017689,"output_tokens":7012589,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_biomysterybench.Bley-WAP.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:46:04.029Z Evidence SHA-256 `59bf79d38ec415a34ffb1c2901a16ecbbe101b7fa864e86104d2571fc7cfc23d` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Arena text style control / non english](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Anon_english.md) | 1,430.67 | — | 54 / 151 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1430.665987401316 \| rating \| reported \| Higher is better \| ### Configuration category non_english config text_style_control ### Reported statistics lower 1422.3484871047328 rank 88 upper 1438.9834876978994 variance 18.008994606679604 votes 5647 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/9818/row` |  |  |  |  |  |  |  |
| [Arena text style control / chinese](https://benchize.com/benchmarks/lmarena%3Atext_style_control%3Achinese.md) | 1,503.31 | — | 38 / 146 | — | — | [Arena](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `rating` \| 1503.308571856167 \| rating \| reported \| Higher is better \| ### Configuration category chinese config text_style_control ### Reported statistics lower 1481.581510549427 rank 52 upper 1525.0356331629068 variance 122.88695911141674 votes 710 ### Source [Original source](https://datasets-server.huggingface.co/parquet/lmarena-ai%2Fleaderboard-dataset/text_style_control/latest) Source Arena Original model label Step 5 Preview Observed 2026-10-21T18:45:05.190Z Evidence SHA-256 `d4875c03cb7ec0c6e8943c3145fd261225e8841b937240dc62a3032c4a798d84` Source locator `/rows/468/row` |  |  |  |  |  |  |  |
| [EMB / misc](https://benchize.com/benchmarks/vals-ai%3Aemb%3Amisc.md) | 49.45% | SE ±9.34 pp | 36 / 47 | $1.12 | 2,643.89source duration | [Vals AI · 1](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 49.445 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 1.12199 \| USD \| per_test \| Lower is better \| \| `latency` \| 2643.893 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±9.34 pp sourcePopulation 76 stderr 9.336 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_emb.BY1LXsAw.json) Source Vals AI Original model label stepfun/step-5-preview Source version 1 Observed 2026-10-21T16:45:57.495Z Evidence SHA-256 `d807aee620ea47152a1fca0257a8dd12a1be4512c76cc1744c1ae9e285180918` Source locator `/tasks/misc/stepfun~1step-5-preview` |  |  |  |  |  |  |  |
| [Finance Agent v2.1 / overall](https://benchize.com/benchmarks/vals-ai%3Afabv2-1%3Aoverall.md) | 58.9% | SE ±2.05 pp | 14 / 18 | $0.58 | 542.51source duration | [Vals AI · 2.1](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.9 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.575115 \| USD \| per_test \| Lower is better \| \| `latency` \| 542.506 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 1024000 provider Stepfun reasoning Not reported Reasoning effort unknown Temperature 1 topP 0.95 verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.05 pp sourcePopulation 28 stderr 2.054 tokenTotals {"cache_read_tokens":214876672,"cache_write_tokens":null,"input_tokens":407803894,"output_tokens":20205838,"reasoning_tokens":0} ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_finance_agent_v2_1.ZbMhy9xo.json) Source Vals AI Original model label stepfun/step-5-preview Source version 2.1 Observed 2026-10-21T16:45:54.954Z Evidence SHA-256 `f1b0c2d7dca4e73b52136862a9be059f7c8481c69f15d705ce90bf3bd9f78429` Source locator `/tasks/overall/stepfun~1step-5-preview` |  |  |  |  |  |  |  |

[Next page](https://benchize.com/models/step-5-preview.md?cursor=6015ad4c8035ee1a88d301dd858b00eef495ff5e9c6a1bcf1661b4e48e40caa7)

---
Cite as: Step 5 Preview — Benchize, https://benchize.com/models/step-5-preview (retrieved 2026-10-21)