mistralai
Mistral: Voxtral Small 24B 2507
mistralai/voxtral-small-24b-2507
Our take
AI-assisted analysis based on the sources below. Suitability is an editorial assessment; test it on your own workload.
Voxtral Small 24B 2507 is Mistral's audio-capable variant built on Mistral Small 3, accepting text, file, and audio input while producing text output. Its model card describes a dedicated transcription mode with automatic language detection, direct audio question answering and summarization, and experimental voice-triggered function calling across eight languages. Within a 32k-token context it reportedly handles roughly 30 minutes of audio for transcription or 40 for understanding. Vendor-reported word error rates appear only as images, so independent validation on target workloads is advisable.[1][2]
Strengths
- Combines speech transcription, audio understanding, and text chat in a single model, so audio Q&A and summaries reportedly need no separate ASR stage.[2]
- Described as natively multilingual across eight languages with automatic language detection in transcription mode.[2]
- Supports tool calling, structured outputs, and a 32k-token context covering up to tens of minutes of audio, per the catalog and card.[1][2]
Limitations
- The card states system prompts are not yet supported and describes function calling as experimental.[2]
- Output is text only, the knowledge cutoff is unstated, and benchmark results are shown only as images, so figures could not be independently verified here.[1][2]
- Self-hosting reportedly requires around 55 GB of GPU memory in bf16/fp16; validate accuracy and cost on your own audio before adoption.[2]
Where to use it
- Multilingual transcription of long recordings such as meetings or interviews, given the dedicated transcription mode; test accuracy on your target domains first.[2]
- Voice assistants that answer questions about or summarize audio content directly, potentially simplifying pipelines that would otherwise chain ASR and LLM components.[2]
- Spoken-intent workflows that trigger backend functions or APIs; promising on paper but flagged experimental, so pilot carefully before production use.[2]
Sources
- [1] OpenRouter catalog · Mistral: Voxtral Small 24B 2507Sources checked: Nov 13, 2026
- [2] Model documentation · Mistral: Voxtral Small 24B 2507Sources checked: Nov 13, 2026
About this model
Description below is the original text supplied by OpenRouter, shown in the source language.
Voxtral Small is an enhancement of Mistral Small 3, incorporating state-of-the-art audio input capabilities while retaining best-in-class text performance. It excels at speech transcription, translation and audio understanding. Input audio...
Specifications
- Model ID
- mistralai/voxtral-small-24b-2507
- Creator
- mistralai
- Added to OpenRouter
- 30.10.2025
- Context length
- 32,768
- Modality
- text+file+audio->text
- Input modalities
- text, audio, file
- Output modalities
- text
- Token pricing
- USD per 1M tokens: in $0.100 / out $0.300
- Knowledge cutoff
- —
- 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, tool_choice, tools |
| Writing | No measurement | output: text |
| Reasoning | No measurement | — |
| Document analysis | No measurement | input: file |
| Vision | No measurement | — |
| Image generation | No measurement | — |
| Audio | No measurement | input: audio |
| 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) |
|---|---|---|---|---|---|---|
| Mistral | 32,768 | $0.100 | $0.300 | — | — | 100% |
| Mistral | 32,000 | $0.110 | $0.330 | — | — | 100% |
Provider data retrieved: 13.11.2026
Related articles
No published articles identify this model yet.