Side-by-side technical and economic comparison between Mistral AI's Codestral (Current Generation) and DeepSeek's DeepSeek-V3 (Legacy API Alias).
• DeepSeek-V3 (Legacy API Alias): Classified as a Legacy Reference model (Legacy Reference (API Aliased)).
| Metric | Codestral (Current Generation) | DeepSeek-V3 (Legacy API Alias) | Advantage |
|---|---|---|---|
| Provider Organization | Mistral AI | DeepSeek | — |
| Standard Input / 1M | $0.30 | $0.14 | DeepSeek-V3 (Legacy API Alias) (53% lower) |
| Cached Input / 1M | $0.03 | $0.014 | DeepSeek-V3 (Legacy API Alias) lower |
| Output / 1M | $0.90 | $0.28 | DeepSeek-V3 (Legacy API Alias) lower |
| Context Window | 256k | 128k | Codestral (Current Generation) (256k) |
| Max Generation Tokens | 8.2k | 8.2k | Codestral (Current Generation) |
| Tokenizer Family | Mistral Tekken BPE (~131k vocabulary) | DeepSeek Byte-level BPE (~128k vocabulary) | — |
Opt for Codestral (Current Generation) when your engineering requirements prioritize Mistral AI's ecosystem, specific tokenizer efficiencies (Exact Tekken Tokenizer (±2%)), or when your expected prompt-to-completion ratios favor its $0.3/M input rate.
Opt for DeepSeek-V3 (Legacy API Alias) when looking for DeepSeek's tooling integration, specific context window depth (128k tokens), or when output generation volume favors its $0.28/M rate.