mistralai
Mistral: Mistral Small 3
mistralai/mistral-small-24b-instruct-2501
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Mistral Small 3 (2501), identified as mistralai/mistral-small-24b-instruct-2501, is Mistral AI's instruction-tuned 24B text-to-text model. The supplied catalog and model card describe a 32K context window, multilingual support, system-prompt handling, function calling, and structured JSON output. Mistral positions it for low-latency assistants, agent workflows, fine-tuning, and local or server deployment. Provider-supplied evaluations are included, but the evidence pack does not contain independent validation, so prospective users should test it on their own tasks.[1][2]
Strengths
- Mistral describes the variant as an instruction-tuned version of Mistral-Small-24B-Base-2501, with text input and output and a 32K-token context window.[1][2]
- The model card reports broad multilingual coverage and strong system-prompt adherence; these are vendor-described capabilities rather than independently verified measurements.[2]
- OpenRouter lists structured output parameters, while Mistral documents function calling through vLLM, supporting tool-oriented integrations.[1][2]
Limitations
- The supplied evidence is limited to an OpenRouter catalog record and Mistral-authored documentation; no independent model card or third-party evaluation is included.[1][2]
- The catalog identifies text as the only input and output modality; support for other modalities is undocumented in this source pack.[1]
- Provider benchmark tables show mixed results against selected peers, and the card says its numbers came from an internal pipeline; validate quality on the intended workload.[2]
Where to use it
- Worth evaluating for multilingual chat or assistant features where low latency is a priority.[2]
- Candidate for prototypes that use function calls or structured JSON, provided the surrounding tool schema is tested end to end.[1][2]
- Potential fit for self-hosted or privacy-conscious deployments and domain fine-tuning, subject to local capacity and fine-tuning validation.[2]
Sources
- [1] OpenRouter catalog · Mistral: Mistral Small 3Sources checked: Nov 13, 2026
- [2] Model documentation · Mistral: Mistral Small 3Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Mistral Small 3 is a 24B-parameter language model optimized for low-latency performance across common AI tasks. Released under the Apache 2.0 license, it features both pre-trained and instruction-tuned versions designed...
Specifications
- Model ID
- mistralai/mistral-small-24b-instruct-2501
- Creator
- mistralai
- Added to OpenRouter
- 30.01.2025
- Context length
- 32,768
- Modality
- text->text
- Input modalities
- text
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.050 / out $0.080
- Knowledge cutoff
- 31.10.2023
- Expiration date
- —
Tasks and capabilities
A grade appears only where the source supplies a matching measurement for that task, the model is active and the cohort holds at least 10 comparable models. The capability column lists facts the source declares — supported formats, reasoning and tool parameters — which are not measured quality.
| Task | Grade | Source-declared capability |
|---|---|---|
| General quality | No measurement | — |
| Coding | No measurement | — |
| Agents | No measurement | parameters: structured_outputs |
| Writing | No measurement | output: text |
| Reasoning | No measurement | — |
| Document analysis | No measurement | — |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | — |
| Video | No measurement | — |
| Embeddings | No measurement | — |
Evaluations
No evaluations are supplied for this model.
How grades work
Each letter grade compares a model only against other models scored by the same source and metric in the current snapshot. S means the top 10%, A the next 20%, B the next 30%, C the rest. Models without a published measurement are shown as not rated — a supported format alone never implies measured quality.
- S — S — top 10% of the cohort
- A — A — next 20% of the cohort
- B — B — next 30% of the cohort
- C — C — below the B band
Providers
Optional live provider statistics from OpenRouter. Missing values are unknown, not zero.
| Provider | Context length | Price in /1M | Price out /1M | Latency p50 (30m, s) | Throughput p50 (30m, tok/s) | Uptime (30m) |
|---|---|---|---|---|---|---|
| DeepInfrafp8 | 32,768 | $0.050 | $0.080 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.