Side-by-side technical and economic comparison between Cohere's Command R (08-2024) and Mistral AI's Codestral (Current Generation).
• Command R (08-2024): Classified as a Legacy Reference model (Tracked Production / Legacy Family).
| Metric | Command R (08-2024) | Codestral (Current Generation) | Advantage |
|---|---|---|---|
| Provider Organization | Cohere | Mistral AI | — |
| Standard Input / 1M | $0.15 | $0.30 | Command R (08-2024) (50% lower) |
| Cached Input / 1M | $0.015 | $0.03 | Command R (08-2024) lower |
| Output / 1M | $0.60 | $0.90 | Command R (08-2024) lower |
| Context Window | 128k | 256k | Codestral (Current Generation) (256k) |
| Max Generation Tokens | 4.1k | 8.2k | Codestral (Current Generation) |
| Tokenizer Family | Cohere BPE (~256k vocabulary) | Mistral Tekken BPE (~131k vocabulary) | — |
Opt for Command R (08-2024) when your engineering requirements prioritize Cohere's ecosystem, specific tokenizer efficiencies (Calibrated Cohere Tokenizer (±3%)), or when your expected prompt-to-completion ratios favor its $0.15/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.