Side-by-side comparison of Codestral 22B (Mistral AI · France) and Gemma 4 31B Instruct (Google DeepMind · United States) for self-hosted deployment of the open-weight model. Codestral 22B is rated conditional; Gemma 4 31B Instruct is conditional. They part ways on licence: Codestral 22B is "MNPL (non-prod)", Gemma 4 31B Instruct 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 Based on published licence terms, Gemma 4 31B ships under pure Apache 2.0 — a notable break from the Gemma Terms of Use used in prior generations — with no prohibited-use carve-outs. US origin carries Schrems-II and CLOUD-Act exposure, and at 30.7B dense the model likely crosses EU AI Act GPAI systemic-risk thresholds that enterprise deployers should document. |
| Last reviewed | 2026-04-15 | 2026-04-17 |
| Open-weight | ||
| Licence | MNPL (non-prod) | Apache 2.0 |
| Commercial use | Paid licence req. | Unrestricted |
| Training data | Partial | Domain-level summary |
| Origin | EU | United States |
| Performance & pricing? | ||
| Quality index | — | 32/100 |
| Speed | — | — |
| Blended price | — | — |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.