# GLM 5.3Max

> Source: Benchize (https://benchize.com) — https://benchize.com/models/glm-5-3-max. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/models/glm-5-3-max.

Z.ai

Canonical URL: https://benchize.com/models/glm-5-3-max
Last updated: Thu, 22 Oct 2026 04:30:12 GMT

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

Z.ai

## Benchmarks

| Benchmark | Score | ± | Rank | Cost | Latency | Source |  |
| --- | --- | --- | --- | --- | --- | --- | --- |
| [Terminal-Bench 4](https://benchize.com/benchmarks/harbor%3Aterminal-bench-4.md) | 41.82% | 95% CI 38.59–45.05% | 8 / 17 | — | — | [harbor](https://www.tbench.ai/) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 41.82 \| percent \| reported \| Higher is better \| \| `pass_at_2` \| 0.5076 \| score_fraction \| reported \| Higher is better \| \| `pass_at_3` \| 0.547 \| score_fraction \| reported \| Higher is better \| \| `pass_at_4` \| 0.5667 \| score_fraction \| reported \| Higher is better \| \| `pass_at_5` \| 0.5758 \| score_fraction \| reported \| Higher is better \| ### Configuration agent Claude Code Reasoning effort max sourceSubmission d72e8775-f4f1-4313-99e7-35b1cb499f24 ### Reported statistics Uncertainty 95% confidence interval: 38.59–45.05% confidenceHalfWidth95 3.23 sampleSize 330 ### Source [Original source](https://www.tbench.ai/) Source harbor Original model label GLM-5.3 Evaluated 2026-08-14 Observed 2026-10-22T04:00:05.388Z Evidence SHA-256 `f02bf923eb5a5a2fae5b1e9f0a0c9ff5fe0e3779f0324a9a4f0e9ed6e322db82` Source locator `/rows/19` |  |  |  |  |  |  |  |
| [FrontierCode 1.1 main](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Amain.md) | 40.14% | — | 69 / 125 | $16.91 | — | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.4014 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.447 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 16.9089 \| USD \| per_rollout \| Lower is better \| ### Configuration harness chisel Reasoning effort max sourceVersion v1_1 subset main ### Reported statistics flaggedRate 0 sampleSize 100 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GLM 5.3 Observed 2026-10-21T16:31:00.878Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/GLM%205.3/max/main` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aoverall.md) | 8.1% | SE ±1.88 pp | 6 / 9 | $2,733.16 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 8.095238095238095 \| percent \| reported \| Higher is better \| \| `total_cost` \| 2733.16185912 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 8486097200 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain overall Reasoning effort max ### Reported statistics Uncertainty Standard error: ±1.88 pp accuracyStandardError 1.8822364227102866 passes 17 rank 12 tasks 210 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics` |  |  |  |  |  |  |  |
| [ToneBench Scriptwriting](https://benchize.com/benchmarks/tonebench%3Ascripts.md) | 2,194.9 | Range 2,163.4–2,230.6 | 16 / 126 | — | — | [tonebench](https://benchmark.towardsai.com/data/leaderboard.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `elo` \| 2194.9 \| rating \| reported \| Higher is better \| \| `overall` \| 82.39 \| percent \| reported \| Higher is better \| \| `anti_slop_numbers` \| 70.42 \| source_score \| rubric \| Higher is better \| \| `continuity_emotion` \| 82.06 \| source_score \| rubric \| Higher is better \| \| `cue_quality` \| 79.69 \| source_score \| rubric \| Higher is better \| \| `hook_strength` \| 83.43 \| source_score \| rubric \| Higher is better \| \| `length_adherence` \| 94.26 \| source_score \| rubric \| Higher is better \| \| `slop` \| 83.4 \| source_score \| rubric \| Higher is better \| \| `slop_score` \| 96.38 \| source_score \| rubric \| Higher is better \| \| `substance_accuracy` \| 83.24 \| source_score \| rubric \| Higher is better \| \| `tone_voice` \| 79.7 \| source_score \| rubric \| Higher is better \| \| `writing_quality` \| 78.67 \| source_score \| rubric \| Higher is better \| \| `youtube_best_practices` \| 81.68 \| source_score \| rubric \| Higher is better \| ### Configuration provider openrouter Reasoning effort unknown scoreScope 10_task_current sourceConfiguration glm-5-3-or thinkingEnabled true ### Reported statistics Uncertainty Reported interval: 2,163.4–2,230.6 interval [2163.4,2230.6] rank 16 runCount 5 ### Source [Original source](https://benchmark.towardsai.com/data/leaderboard.json) Source tonebench Original model label GLM-5.3 (max) Evaluated 2026-10-08 Observed 2026-10-21T16:32:48.862Z Evidence SHA-256 `4264338337f418d3ded7ace250cad64fe12cacbd4b79a312bfaf0ae9268a01c3` Source locator `/rows/15` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · mathematical](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Amathematical.md) | 5.88% | SE ±3.29 pp | 6 / 9 | $1,571.92 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 5.88235294117647 \| percent \| reported \| Higher is better \| \| `total_cost` \| 1571.9234520000002 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 2053648887 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain mathematical Reasoning effort max ### Reported statistics Uncertainty Standard error: ±3.29 pp accuracyStandardError 3.294776668301199 passes 3 rank 12 tasks 51 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics/domain_metrics/mathematical` |  |  |  |  |  |  |  |
| [BullshitBench v2](https://benchize.com/benchmarks/bullshitbench%3Av2.md) | 66% | — | 30 / 191 | — | — | [bullshitbench · v2](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `green_rate` \| 0.66 \| score_fraction \| all_attempts \| Higher is better \| \| `red_rate` \| 0.1 \| score_fraction \| all_attempts \| Lower is better \| \| `refusal_rate` \| 0.0 \| score_fraction \| all_attempts \| Lower is better \| \| `avg_score` \| 1.5433 \| score \| judge_grade_0_to_2 \| Higher is better \| ### Configuration Reasoning effort max suite v2 thinkingEnabled true ### Reported statistics acceptedCount 10 answeredCount 100 clearCount 66 errorCount 0 partialCount 24 rank 37 refusalCount 0 sampleSize 100 ### Source [Original source](https://api.github.com/repos/petergpt/bullshit-benchmark/contents/data/v2/latest/leaderboard.csv?ref=3c655234d15967738f95da696e3f938d4862dcf7) Source bullshitbench Original model label z-ai/glm-5.3 Source version v2 Observed 2026-10-22T03:10:12.603Z Evidence SHA-256 `98b485e08764d1c97144bc4f7b28ced1a96d50850947c775870dfda0e5d749b5` Source locator `/rows/36` |  |  |  |  |  |  |  |
| [NC Bench](https://benchize.com/benchmarks/nc-bench%3Aoverall.md) | 95.2% | — | 5 / 144 | — | — | [nc bench](https://www.nc-bench.com/models) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 95.2 \| percent \| reported \| Higher is better \| \| `tooling` \| 100 \| percent \| category \| Higher is better \| \| `creative_writing` \| 89.99 \| percent \| category \| Higher is better \| \| `language` \| 96.87 \| percent \| category \| Higher is better \| \| `utility` \| 98.44 \| percent \| category \| Higher is better \| \| `reasoning` \| 93.04 \| percent \| category \| Higher is better \| \| `text_editing` \| 98.75 \| percent \| category \| Higher is better \| \| `rule_following` \| 89.17 \| percent \| category \| Higher is better \| \| `hallucination` \| 95.3 \| percent \| category \| Higher is better \| ### Configuration Reasoning effort max Model configuration Z.AI GLM 5.3 (Reasoning, Max) thinkingEnabled true ### Source [Original source](https://www.nc-bench.com/models) Source nc bench Original model label Z.AI GLM 5.3 (Reasoning, Max) Observed 2026-10-21T16:31:48.370Z Evidence SHA-256 `ec261022458543624b08a136d01f5a2f0666cd52ffb49c64d6ca983e8543d1b5` Source locator `/rows/6` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · earth](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aearth.md) | 4.17% | SE ±4.08 pp | 6 / 9 | $1,404.89 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 4.166666666666666 \| percent \| reported \| Higher is better \| \| `total_cost` \| 1404.890923 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 1588775692 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain earth Reasoning effort max ### Reported statistics Uncertainty Standard error: ±4.08 pp accuracyStandardError 4.078937543219003 passes 1 rank 12 tasks 24 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics/domain_metrics/earth` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · life](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Alife.md) | 10.53% | SE ±4.06 pp | 4 / 9 | $1,310.93 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 10.526315789473683 \| percent \| reported \| Higher is better \| \| `total_cost` \| 1310.9276260000004 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 1506421392 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain life Reasoning effort max ### Reported statistics Uncertainty Standard error: ±4.06 pp accuracyStandardError 4.064886556427542 passes 6 rank 12 tasks 57 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics/domain_metrics/life` |  |  |  |  |  |  |  |
| [DeepSWE](https://benchize.com/benchmarks/deepswe%3Av1.1.md) | 68.96% | Range 65.94–71.98% | 16 / 70 | $3.99 | 2,116.18s | [deepswe](https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `pass_at_1` \| 0.6895787139689579 \| score_fraction \| scored_rollout_attempts \| Higher is better \| \| `pass_at_4` \| 0.8761061946902655 \| score_fraction \| tasks_with_one_passing_rollout \| Higher is better \| \| `cost_per_attempt` \| 3.9933584893126386 \| USD \| scored_attempt \| Lower is better \| \| `duration` \| 2116.1784804168515 \| seconds \| scored_attempt \| Lower is better \| ### Configuration harness mini-swe-agent Reasoning effort max sourceConfiguration mini_swe_agent_glm_5_3_max ### Reported statistics Uncertainty Reported interval: 65.94–71.98% interval [0.6593922012533708,0.7197652266845449] intervalMethod 95% run-to-run: SE across repeated whole-benchmark passes (1.96 * std(runs)/sqrt(R)) runCount 4 sampleSize 451 taskCount 113 ### Source [Original source](https://deepswe.datacurve.ai/artifacts/v1.1/leaderboard-live.json) Source deepswe Original model label glm-5-3 Observed 2026-10-22T03:50:05.889Z Evidence SHA-256 `a7c15d66288fd249c020b9931c017b92d1a3b90e480b3ff34974b752bd030019` Source locator `/rows/15` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · physical](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aphysical.md) | 9.8% | SE ±4.16 pp | 5 / 9 | $1,726.56 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 9.803921568627452 \| percent \| reported \| Higher is better \| \| `total_cost` \| 1726.556125 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 2345166065 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain physical Reasoning effort max ### Reported statistics Uncertainty Standard error: ±4.16 pp accuracyStandardError 4.163980181949191 passes 5 rank 12 tasks 51 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics/domain_metrics/physical` |  |  |  |  |  |  |  |
| [Short Story Writing — Pairwise Comparison](https://benchize.com/benchmarks/lechmazur-short-story%3Apairwise.md) | 3.03 | 95% CI 2.94–3.11 | 6 / 50 | — | — | [lechmazur short story · 8ba548d1cffefa5c472d3832e05f64e119a88fa4](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `thurstone_rating` \| 3.0277077443758276 \| index \| pooled_observed_panel \| Higher is better \| \| `win_probability` \| 0.8804767850780137 \| score_fraction \| comparison_pool \| Higher is better \| ### Configuration benchmarkScope data/writing_gemini_argon_20261013 Reasoning effort max thinkingEnabled true ### Reported statistics Uncertainty 95% confidence interval: 2.94–3.11 bootstrapSamples 300 confidenceInterval95 ["2.9420163968638193","3.107084573676175"] sourceRank 6 ### Source [Original source](https://api.github.com/repos/lechmazur/writing/contents/data/writing_gemini_argon_20261013/ratings.csv?ref=8ba548d1cffefa5c472d3832e05f64e119a88fa4) Source lechmazur short story Original model label GLM-5.3 (max) Source version 8ba548d1cffefa5c472d3832e05f64e119a88fa4 Observed 2026-10-21T16:34:21.130Z Evidence SHA-256 `c36baebed9f302ff7ad8fe97907d72e1757a015a39b6a3b23437597cf8e3bff6` Source locator `/rows/5` |  |  |  |  |  |  |  |
| [FrontierCode 1.1 extended](https://benchize.com/benchmarks/frontiercode%3Av1_1%3Aextended.md) | 52.63% | — | 76 / 125 | $13.79 | — | [frontiercode](https://cognition.com/data/frontiercode-leaderboard/data.json) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 0.5263 \| score_fraction \| blocking_criteria_adjusted \| Higher is better \| \| `correct` \| 0.5773 \| score_fraction \| raw_rubric \| Higher is better \| \| `cost` \| 13.7873 \| USD \| per_rollout \| Lower is better \| ### Configuration harness chisel Reasoning effort max sourceVersion v1_1 subset extended ### Reported statistics flaggedRate 0 sampleSize 150 ### Source [Original source](https://cognition.com/data/frontiercode-leaderboard/data.json) Source frontiercode Original model label GLM 5.3 Observed 2026-10-21T16:31:00.878Z Evidence SHA-256 `ff0040778b7bdf69c705ad7474f771fb5e53cfaf5cf8573b3a4614153543d75f` Source locator `/v1_1/data/GLM%205.3/max/extended` |  |  |  |  |  |  |  |
| [Terminal-Bench Science 0.1 · engineering](https://benchize.com/benchmarks/terminal-science%3Av0-1-eval%3Aengineering.md) | 7.41% | SE ±5.04 pp | 6 / 9 | $788.27 | — | [terminal science](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `accuracy` \| 7.4074074074074066 \| percent \| reported \| Higher is better \| \| `total_cost` \| 788.2680359999999 \| USD \| all_trials \| Lower is better \| \| `total_tokens` \| 992085164 \| tokens \| all_trials \| Lower is better \| ### Configuration agent Claude Code benchmarkVersion v0-1-eval domain engineering Reasoning effort max ### Reported statistics Uncertainty Standard error: ±5.04 pp accuracyStandardError 5.040102351405716 passes 2 rank 12 tasks 27 ### Source [Original source](https://www.terminal-bench-science.ai/api/leaderboard?package=terminal-bench-science%2Fterminal-bench-science&name=v0-1-eval) Source terminal science Original model label GLM 5.3 Evaluated 2026-09-16T13:45:27.369938+00:00 Observed 2026-10-21T16:32:46.684Z Evidence SHA-256 `86e6278ca64da88b69c1efd6059a33248bd300f27673098f60184286b470c4a7` Source locator `/rows/11/metrics/domain_metrics/engineering` |  |  |  |  |  |  |  |

---
Cite as: GLM 5.3Max — Benchize, https://benchize.com/models/glm-5-3-max (retrieved 2026-10-22)