Side-by-side comparison of GLM-4.5 (Zhipu AI · China) and Mistral Large 2 (Mistral AI · France) for self-hosted deployment of the open-weight model. GLM-4.5 is rated conditional; Mistral Large 2 is conditional. They part ways on licence: GLM-4.5 is "MIT", Mistral Large 2 is "MRL (research)".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | Conditional MoE flagship from Zhipu under MIT. Strong agentic and coding benchmarks. Same Chinese-origin alignment and geopolitical considerations as DeepSeek / Qwen. | Conditional Frontier 123B dense model from an EU vendor. Open weights under Mistral Research License — non-commercial by default. Commercial deployment requires a separate paid licence from Mistral AI. |
| Last reviewed | 2026-04-15 | 2026-04-15 |
| Open-weight | ||
| Licence | MIT | MRL (research) |
| Commercial use | Yes | Paid licence req. |
| Training data | Undisclosed | Undisclosed |
| Origin | China | EU |
| Performance & pricing? | ||
| Quality index | 26/100 | 15/100 |
| Speed | 44 tok/s | 38 tok/s |
| Blended price | $0.84/M | $3.00/M |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.