# GPT-3.5 TurboHigh > Source: Benchize (https://benchize.com) — https://benchize.com/models/gpt-3-5-turbo-high. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/gpt-3-5-turbo-high. OpenAI Canonical URL: https://benchize.com/models/gpt-3-5-turbo-high Last updated: Thu, 22 Oct 2026 08:20:15 GMT Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations. OpenAI ## Benchmarks | Benchmark | Score | ± | Rank | Cost | Latency | | | --- | --- | --- | --- | --- | --- | --- | | [LegalBench / rhetoric tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Arhetoric_tasks.md) | 65.71% | SE ±1.1 pp | 43 / 44 | — | 0.47source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.714 \| percent \| reported \| Higher is better \| \| `latency` \| 0.465 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.1 pp sourcePopulation 151 stderr 1.095 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/rhetoric_tasks/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / hispanic](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Ahispanic.md) | 58% | SE ±2.16 pp | 12 / 12 | $0.000426 | 1.5source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000426 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.495 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 95 stderr 2.161 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/hispanic/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / overall](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Aoverall.md) | 58.47% | SE ±0.88 pp | 12 / 12 | $0.000424 | 1.59source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.471 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000424 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.589 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.88 pp sourcePopulation 95 stderr 0.881 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/overall/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / asian](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Aasian.md) | 58.6% | SE ±2.16 pp | 12 / 12 | $0.000424 | 1.32source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.6 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000424 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.324 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 95 stderr 2.157 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/asian/openai~1gpt-3.5-turbo` | | | | | | | | [LegalBench / interpretation tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Ainterpretation_tasks.md) | 65.52% | SE ±0.21 pp | 43 / 44 | — | 0.46source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 65.515 \| percent \| reported \| Higher is better \| \| `latency` \| 0.46 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.21 pp sourcePopulation 151 stderr 0.21 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/interpretation_tasks/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / unbiased](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Aunbiased.md) | 59.25% | SE ±2.15 pp | 12 / 12 | $0.000415 | 1.59source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 59.25 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000415 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.588 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.15 pp sourcePopulation 95 stderr 2.152 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/unbiased/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / black](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Ablack.md) | 58.73% | SE ±2.16 pp | 12 / 12 | $0.000428 | 1.77source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.726 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000428 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.771 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 95 stderr 2.159 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/black/openai~1gpt-3.5-turbo` | | | | | | | | [GPQA Diamond / diamond few shot cot](https://benchize.com/benchmarks/vals-ai%3Agpqa%3Adiamond_few_shot_cot.md) | 28.79% | SE ±3.2 pp | 39 / 40 | $0.0013 | 2.83source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 28.788 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.001296 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.834 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.2 pp sourcePopulation 138 stderr 3.201 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_gpqa.ghZZBpyU.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:20.661Z Evidence SHA-256 `951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077` Source locator `/tasks/diamond_few_shot_cot/openai~1gpt-3.5-turbo` | | | | | | | | [GPQA Diamond](https://benchize.com/benchmarks/gpqa-diamond.md) | 30.56% | SE ±2.29 pp | 39 / 40 | $0.000862 | 2.63source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 30.556 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000862 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.628 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.29 pp sourcePopulation 138 stderr 2.287 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_gpqa.ghZZBpyU.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:20.661Z Evidence SHA-256 `951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077` Source locator `/tasks/overall/openai~1gpt-3.5-turbo` | | | | | | | | [LegalBench / overall](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aoverall.md) | 64.37% | SE ±0.63 pp | 43 / 44 | — | 0.54source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 64.372 \| percent \| reported \| Higher is better \| \| `latency` \| 0.542 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.63 pp sourcePopulation 151 stderr 0.628 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/overall/openai~1gpt-3.5-turbo` | | | | | | | | [GPQA Diamond / diamond zero shot cot](https://benchize.com/benchmarks/vals-ai%3Agpqa%3Adiamond_zero_shot_cot.md) | 32.32% | SE ±3.27 pp | 40 / 40 | $0.000427 | 2.42source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 32.323 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000427 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.423 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±3.27 pp sourcePopulation 138 stderr 3.266 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_gpqa.ghZZBpyU.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:20.661Z Evidence SHA-256 `951eb5224097733a9d09d06c14908ae2980e2e640e539f2c284877c5092d7077` Source locator `/tasks/diamond_zero_shot_cot/openai~1gpt-3.5-turbo` | | | | | | | | [LegalBench / rule tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Arule_tasks.md) | 66.17% | SE ±2.66 pp | 41 / 44 | — | 0.78source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 66.174 \| percent \| reported \| Higher is better \| \| `latency` \| 0.777 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.66 pp sourcePopulation 151 stderr 2.659 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/rule_tasks/openai~1gpt-3.5-turbo` | | | | | | | | [LegalBench / issue tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aissue_tasks.md) | 57.05% | SE ±0.52 pp | 43 / 44 | — | 0.67source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 57.054 \| percent \| reported \| Higher is better \| \| `latency` \| 0.665 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±0.52 pp sourcePopulation 151 stderr 0.515 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/issue_tasks/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / white](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Awhite.md) | 58.1% | SE ±2.16 pp | 12 / 12 | $0.000426 | 2source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.1 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000426 \| USD \| per_test \| Lower is better \| \| `latency` \| 2.002 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 95 stderr 2.16 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/white/openai~1gpt-3.5-turbo` | | | | | | | | [MedQA / indigenous](https://benchize.com/benchmarks/vals-ai%3Amedqa%3Aindigenous.md) | 58.15% | SE ±2.16 pp | 12 / 12 | $0.000428 | 1.36source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 58.15 \| percent \| reported \| Higher is better \| \| `cost_per_test` \| 0.000428 \| USD \| per_test \| Lower is better \| \| `latency` \| 1.356 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±2.16 pp sourcePopulation 95 stderr 2.16 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_medqa.Cco1kvog.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:15.908Z Evidence SHA-256 `d67905978d932f64698ebe9e230d9519e5bae1c9d7176dda8379dc3874b6f90f` Source locator `/tasks/indigenous/openai~1gpt-3.5-turbo` | | | | | | | | [LegalBench / conclusion tasks](https://benchize.com/benchmarks/vals-ai%3Alegal_bench%3Aconclusion_tasks.md) | 67.4% | SE ±1.13 pp | 44 / 44 | — | 0.35source duration | | | ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 67.401 \| percent \| reported \| Higher is better \| \| `latency` \| 0.345 \| source_duration \| reported_duration \| Lower is better \| ### Configuration computeEffort Not reported harness Not reported maxOutputTokens 4096 provider OpenAI reasoning Not reported Reasoning effort high Temperature 1 topP Not reported verbosity Not reported ### Reported statistics Uncertainty Standard error: ±1.13 pp sourcePopulation 151 stderr 1.13 tokenTotals Not reported ### Source [Original source](https://www.vals.ai/_astro/benchmark_view_legal_bench.D6uTgA3b.json) Source Vals AI Original model label openai/gpt-3.5-turbo Source version 1 Observed 2026-10-22T02:10:08.909Z Evidence SHA-256 `c8486ba756baa5cd3955d035edff67440e8db23e69d62ad1a9e18d15b8c4bc52` Source locator `/tasks/conclusion_tasks/openai~1gpt-3.5-turbo` | | | | | | | --- Cite as: GPT-3.5 TurboHigh — Benchize, https://benchize.com/models/gpt-3-5-turbo-high (retrieved 2026-10-22)