Side-by-side technical and economic comparison between Mistral AI's Mistral Large 3 and Meta / Open's Llama 4 Scout (109B).
| Metric | Mistral Large 3 | Llama 4 Scout (109B) | Advantage |
|---|---|---|---|
| Provider Organization | Mistral AI | Meta / Open | — |
| Standard Input / 1M | $0.50 | $0.30 | Llama 4 Scout (109B) (40% lower) |
| Cached Input / 1M | $0.10 | $0.30 | Mistral Large 3 lower |
| Output / 1M | $1.50 | $0.60 | Llama 4 Scout (109B) lower |
| Context Window | 128k | 10M | Llama 4 Scout (109B) (10M) |
| Max Generation Tokens | 16.4k | 16.4k | Mistral Large 3 |
| Tokenizer Family | Mistral Tekken Tokenizer (131k vocabulary) | Meta Llama 3/4 Tiktoken (128k vocabulary) | — |
Opt for Mistral Large 3 when your engineering requirements prioritize Mistral AI's ecosystem, specific tokenizer efficiencies (Calibrated Mistral Tokenizer (±3%)), or when your expected prompt-to-completion ratios favor its $0.5/M input rate.
Opt for Llama 4 Scout (109B) when looking for Meta / Open's tooling integration, specific context window depth (10M tokens), or when output generation volume favors its $0.6/M rate.