Side-by-side comparison of Codestral 22B (Mistral AI · France) and Grok-2 (xAI · USA) for self-hosted deployment of the open-weight model. Codestral 22B is rated conditional; Grok-2 is conditional. They part ways on licence: Codestral 22B is "MNPL (non-prod)", Grok-2 is "Grok 2 Community".
| 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 xAI's first open-weights release. Commercial use allowed under the Grok 2 Community License with xAI's Acceptable Use Policy. Notable restriction: weights cannot be used to train other models (distillation ban). 500GB model, needs 8 GPUs with 40GB+. |
| Last reviewed | 2026-04-15 | 2026-04-16 |
| Open-weight | ||
| Licence | MNPL (non-prod) | Grok 2 Community |
| Commercial use | Paid licence req. | Yes (w/ AUP) |
| Training data | Partial | Undisclosed |
| Origin | EU | USA |
| Performance & pricing? | ||
| Quality index | — | 14/100 |
| Speed | — | — |
| Blended price | — | — |
| Context window | — | — |
| Evidence | ||
| Sources | ||
No overlapping sources between the two entries.