Side-by-side technical and economic comparison between xAI's Grok 2 and Mistral AI's Codestral (Current Generation).
• Grok 2: Classified as a Legacy Reference model (Generally Available).
| Metric | Grok 2 | Codestral (Current Generation) | Advantage |
|---|---|---|---|
| Provider Organization | xAI | Mistral AI | — |
| Standard Input / 1M | $2.00 | $0.30 | Codestral (Current Generation) (85% lower) |
| Cached Input / 1M | None | $0.03 | Codestral (Current Generation) lower |
| Output / 1M | $10.00 | $0.90 | Codestral (Current Generation) lower |
| Context Window | 128k | 256k | Codestral (Current Generation) (256k) |
| Max Generation Tokens | 8.2k | 8.2k | Grok 2 |
| Tokenizer Family | xAI Byte-level BPE (~131k vocabulary) | Mistral Tekken BPE (~131k vocabulary) | — |
Opt for Grok 2 when your engineering requirements prioritize xAI's ecosystem, specific tokenizer efficiencies (Calibrated xAI Tokenizer (±3%)), or when your expected prompt-to-completion ratios favor its $2/M input rate.
Opt for Codestral (Current Generation) when looking for Mistral AI's tooling integration, specific context window depth (256k tokens), or when output generation volume favors its $0.9/M rate.