mistral-embed

mistral-embed is a budget-tier language model from Mistral. mistral-embed is a proprietary model available exclusively through Mistral's API. Mistral's model family is known for strong performance per parameter, with the 7B model punching well above its weight class and the Mixtral MoE variants delivering near-frontier quality at efficient inference cost.

mistral-embed is priced at $0.1/1M input tokens. Output token pricing does not apply to this model — it returns a vector or score rather than generated text. At 1,000 tokens per call, mistral-embed costs approximately $0.10 per 1,000 API calls.

Its 8K-token context window is standard for this tier — adequate for most chat and completion tasks, but may require chunking for long-document workloads. Median latency (P50 time-to-first-token) is approximately 80ms — fast enough for real-time interactive applications.

mistral-embed is best suited to embedding, search, cost. It is a text-only model without vision or function calling support. When evaluating this model for a specific workload, the most important factors are: context window (does your input fit?), output price (how long are your generations?), and latency (is this interactive or batch?). The pricing table above shows how mistral-embed compares to similar models across these dimensions.

mistral-embed sits in the budget tier — the most cost-efficient models, best for high-volume simple tasks, classification, extraction, and applications where cost per call is the primary constraint. Unlike open-weights alternatives, mistral-embed is only available via Mistral's API — there is no self-hosted option, which simplifies deployment but creates vendor dependency. When comparing mistral-embed to alternatives, use the similar models section on this page to find models at comparable price points with different capability tradeoffs. The intelligence score (0/100) reflects aggregate benchmark performance — a useful proxy for general capability, though real-world performance varies significantly by task type. Always test with your specific prompts and use case before committing to a model for production.