Side-by-side comparison of Mistral Small 4 (Mistral AI · France) and Qwen 2.5 (Alibaba · China) for self-hosted deployment of the open-weight model. Mistral Small 4 is rated EU-ready; Qwen 2.5 is conditional. They part ways on origin: Mistral Small 4 is "EU", Qwen 2.5 is "China".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | EU-ready Unified model folding Instruct, reasoning (Magistral) and code (Devstral) into a single 119B MoE under Apache 2.0. 6.5B active params, 256K context, 24 languages, toggleable reasoning effort. Strongest permissive EU option at this scale. | Conditional Legally clean under Apache 2.0, but Chinese origin raises supply-chain and geopolitical questions. Vet carefully for sensitive use cases. |
| Last reviewed | 2026-04-15 | 2026-04-15 |
| Open-weight | ||
| Licence | Apache 2.0 | Apache 2.0 |
| Commercial use | Yes | Yes |
| Training data | Undisclosed | Undisclosed |
| Origin | EU | China |
| Performance & pricing? | ||
| Quality index | 19/100 | 16/100 |
| Speed | 147 tok/s | 55 tok/s |
| Blended price | $0.26/M | — |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.