← comparisons / glm-5.1-vs-mistral-medium-2604
GLM-5.1 vs Mistral Medium 3.5
01Head to head
| GLM-5.1 | Mistral Medium 3.5 | |
|---|---|---|
| Creator | Zhipuai | Mistral AI |
| Released | 2026-04-07 | 2026-04-29 |
| Context | 200K | 262K |
| Max output | 131K | 262K |
| Input $/M | $0.98/M | — |
| Output $/M | $3.08/M | — |
| Knowledge cutoff | — | — |
| Modalities | text → text | text + image → text |
| Capabilities | tool-use, extended-thinking, structured-output | tool-use, extended-thinking, structured-output |
| Providers | OpenRouter | Mistral AI (La Plateforme) |
02Benchmarks · vendor-reported unless verified
| GLM-5.1 | Mistral Medium 3.5 | |
|---|---|---|
| Artificial Analysis Coding Agent Index (average pass@1) | 52.7 | — |
| SWE-Atlas Codebase QnA (pass@1) | 73.2 | — |
| SWE-bench Pro (pass@1) | 19.8 | — |
| SWE-bench Verified (resolved) | — | 77.6 |
| Terminal-Bench (pass@1) | 65.1 | — |
| τ³-Telecom (accuracy) | — | 91.4 |