Updated daily at 05:00 UTC
- Released
- 16 Jun 2026
- Status
- Legacy
- Context
- 1M
- Max output
- 131.1K
- Input per 1M tokens
- $1.40
- Output per 1M tokens
- $4.40
- Open weights
- Yes
- License
- mit
- Parameters
- —
- Knowledge cutoff
- —
- Reasoning
- Yes
- Modalities
- text → text
Category scores
Out of 100, from independent and third-party results only, by method category-2026-09-29.1.
| Category | Score | Rank | Coverage |
|---|---|---|---|
| Knowledge & reasoning | 32.7 | 51 | 4 of 4 |
| Coding & software engineering | 34.1 | Not ranked | 2 of 5 |
| Factuality & hallucination | 34.2 | 45 | 1 of 1 |
| Human preference | 68.5 | Not ranked | 2 of 5 |
Benchmark results
Every result we hold for this model, with who ran it and where it was published.
Knowledge & reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Chess Puzzles | Puzzles solved | 21.0% | Independent | max | 17 Jun 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 14.0% | Independent | low | 10 Aug 2026 | Epoch AI |
| Chess Puzzles | Puzzles solved | 6.0% | Independent | none | 10 Aug 2026 | Epoch AI |
| DTBench (Decision Theory Benchmark) | Accuracy (capability questions) | 93.6% | Third party | max | — | https://conceptualreasoning.ai/dtbench via Epoch AI |
| Earthborne Rangers (EBR-bench) | Objectives reached | 9.5% | Independent | max | 29 Jun 2026 | Epoch AI |
| Humanity's Last Exam (HLE) | Accuracy | 54.7% | Self-reported | — | — | Model Card via Hugging Face Community Evals |
| LMCA (Language Model Conceptual Argumentation) | Correlation with expert ratings (Pearson r) | 0.4582 | Third party | max | — | https://conceptualreasoning.ai/lmca via Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 19.0% | Independent | low | 27 Aug 2026 | Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 18.0% | Independent | minimal | 27 Aug 2026 | Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 18.0% | Independent | none | 27 Aug 2026 | Epoch AI |
| Mystery Game Puzzles | Puzzles solved | 15.0% | Independent | medium | 27 Aug 2026 | Epoch AI |
| SimpleBench | Score (AVG@5) | 58.8% | Third party | unknown | — | https://lmcouncil.ai/benchmarks via Epoch AI |
Scientific reasoning
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| GPQA Diamond | Accuracy | 91.9% | Independent | max | 24 Jun 2026 | Epoch AI |
| GPQA Diamond | Accuracy | 87.9% | Independent | low | 10 Aug 2026 | Epoch AI |
| GPQA Diamond | Accuracy | 71.2% | Independent | none | 10 Aug 2026 | Epoch AI |
| GPQA Diamond | Accuracy | 91.2% | Self-reported | — | — | Model Card via Hugging Face Community Evals |
| Surface Evolver Bench | Mean score (partial credit) | 55.6% | Third party | high | — | https://yhenon.github.io/surface-evolver-llm-eval/ via Epoch AI |
Mathematics
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| FrontierMath Tier 4 | Accuracy | 29.3% | Independent | max | 19 Jun 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 59.2% | Independent | max | 19 Jun 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 54.7% | Independent | low | 29 Aug 2026 | Epoch AI |
| FrontierMath Tiers 1-3 | Accuracy | 42.5% | Independent | none | 29 Aug 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 86.4% | Independent | max | 25 Jun 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 75.6% | Independent | low | 10 Aug 2026 | Epoch AI |
| OTIS Mock AIME 2024-2025 | Accuracy | 28.9% | Independent | none | 10 Aug 2026 | Epoch AI |
Coding & software engineering
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| DeepSWE | Pass@1 | 43.8% | Third party | max | — | https://deepswe.datacurve.ai/ via Epoch AI |
| DeepSWE | Pass@1 | 36.3% | Third party | high | — | https://deepswe.datacurve.ai/ via Epoch AI |
| FrontierCode | Main score (Mean@5) | 24.5% | Third party | none | — | https://cognition.com/frontiercode via Epoch AI |
| PostTrainBench | Average score | 31.7% | Source not stated | max | — | Epoch AI via Epoch AI |
| SWE-bench Verified | % resolved | 78.7% | Independent | max | 25 Jun 2026 | Epoch AI |
| WeirdML (v2) | Average accuracy | 70.1% | Third party | max | — | https://htihle.github.io/weirdml.html via Epoch AI |
| WeirdML (v2) | Average accuracy | 67.3% | Third party | high | — | https://htihle.github.io/weirdml.html via Epoch AI |
Factuality & hallucination
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| SimpleQA Verified | Accuracy (% correct) | 34.2% | Independent | max | 27 Aug 2026 | Epoch AI |
Human preference
| Benchmark | Metric | Value | Provenance | Setting | Evaluated | Source |
|---|---|---|---|---|---|---|
| Code Arena: WebDev (arena.ai) | Arena score (Elo scale) | 1602 | Independent | max | 25 Sep 2026 | Arena |
| Code Arena: WebDev (arena.ai) | Votes | 13601 | Independent | max | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Arena score (Elo scale) | 1476 | Independent | max | 25 Sep 2026 | Arena |
| Text Arena (arena.ai) | Votes | 43570 | Independent | max | 25 Sep 2026 | Arena |
Other GLM versions
- GLM-5.3-Flash26 Aug 2026
- GLM-5.318 Aug 2026
- GLM-5.17 Apr 2026
- GLM-5-Turbo15 Mar 2026
- GLM-512 Feb 2026
- GLM-4.7-Flash19 Jan 2026
- GLM-4.722 Dec 2025
- GLM-4.630 Sep 2025
- GLM-4.528 Jul 2025
- GLM-4.5-Air28 Jul 2025
- GLM-4.5-Flash28 Jul 2025
- GLM-4-32B-041414 Apr 2025
- GLM-4-9B-041414 Apr 2025
- GLM-4-PlusAug 2024
- GLM-4-9B-Chat5 Jun 2024
- GLM-4-9B-Chat-1M5 Jun 2024
- GLM-4Jan 2024
- GLM-4-Flash—
- GLM-4-Long—