GLM-5
GLM-5.1
Z.ai (Zhipu AI) · open-permissive mit
| Parameters (B) | 753.864 |
|---|---|
| Architecture | moe |
| Layers / KV heads / head dim | 78 / 64 / 64 |
| Context window | 202752 |
| Modalities | in: text; out: text |
| GGUF | available |
| Commercial use | yes |
Minimum hardware (estimate)
At Q4_K_M and 8k context, total working set ≈ 468.9 GiB. On a machine with 64 GiB system RAM and no discrete GPU: Won't run. Limiting factor: No GPU and system RAM insufficient for weights + KV.
Scores
| Benchmark | Value | Uncertainty | Run by | Source |
|---|---|---|---|---|
| GPQA Diamond | 85.5% | ±0.0207 | third-party | epoch-ai |
| SWE-bench Verified | 74.2% | ±0.0199 | third-party | epoch-ai |
| OTIS Mock AIME 2024-2025 | 92.2% | ±0.0354 | third-party | epoch-ai |
| FrontierMath | 33.5% | ±0.0278 | third-party | epoch-ai |
| FrontierMath Tier 4 | 12.5% | ±0.0482 | third-party | epoch-ai |
| SimpleQA Verified | 37.3% | ±0.0153 | third-party | epoch-ai |
Provenance
- identity, license, architecture, parameter count: https://huggingface.co/zai-org/GLM-5.1 (retrieved 2026-08-10)
HF: zai-org/GLM-5.1 · snapshot 2026-08-10