Side-by-side comparison of Codestral 22B (Mistral AI · France) and Qwen3.6-27B (Alibaba (Qwen)) for self-hosted deployment of the open-weight model. Codestral 22B is rated conditional; Qwen3.6-27B is conditional. They part ways on licence: Codestral 22B is "MNPL (non-prod)", Qwen3.6-27B is "Apache 2.0".
| Field | ||
|---|---|---|
| Summary | ||
| Verdict | Conditional EU code model trained on 80+ languages. Licensed under Mistral Non-Production License — blocked for any production or commercial deployment without a paid commercial licence. Use Codestral Mamba (Apache 2.0) if you need commercial freedom. | Conditional Per the published Apache 2.0 licence, the Qwen3.6-27B weights are deployable without commercial restriction, including for vision-language and 1M-context workloads. The blockers for regulated EU use are the China-based vendor and the absence of any training-data disclosure on the model card — both should be mitigated through self-hosting and a deployer-prepared GPAI compliance file. |
| Last reviewed | 2026-04-15 | 2026-04-28 |
| Open-weight | ||
| Licence | MNPL (non-prod) | Apache 2.0 |
| Commercial use | Paid licence req. | Unrestricted |
| Training data | Partial | Undisclosed |
| Origin | EU | China |
| Performance & pricing? | ||
| Quality index | — | 46/100 |
| Speed | — | 66 tok/s |
| Blended price | — | $1.35/M |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.