mistralai/Voxtral-Mini-4B-Realtime-2602
mistralai/Voxtral-Mini-4B-Realtime-2602 is a mistralai record on this register. The Hugging Face model API lists the pipeline tag as automatic-speech-recognition. The safetensors index gives an exact parameter count of 4,429,679,360. Architectural context length (max_position_embeddings) is 1,500. The listed licence is apache-2.0. The Hub repository was created on 21 Jan 2026, which is the release date this register uses.
Primary source: the Hugging Face model card for mistralai/Voxtral-Mini-4B-Realtime-2602. Fields below come from that card, the Hub model API, and the repository’s own config.json when those files exist. A field we cannot source is omitted rather than guessed.
| Organisation | mistralai |
|---|---|
| ReleasedRepo creation date on Hugging Face | 21 Jan 2026 |
| Last modified | 11 Mar 2026 |
| ParametersExact, from the safetensors index | 4.4B 4,429,679,360 |
| Context lengthmax_position_embeddings — architectural maximum, not necessarily served | 1K 1,500 |
| Licence | apache-2.0 |
| Pipeline / taskpipeline_tag on the Hugging Face model API | automatic-speech-recognition |
| Base modelDeclared on the Hub; not a page on this register | mistralai/Ministral-3-3B-Base-2512 |
| File formats | safetensors |
| Librarylibrary_name on the Hugging Face model API | vllm |
| Architecture | voxtral_realtime |
| Layers | 32 |
| Vocabulary | 131,072 |
| DownloadsAt last check | 2.1m |
| Likes | 1k |
| Gated | No |
Lineage
The Hub card names mistralai/Ministral-3-3B-Base-2512 as the base model. That id is not a page on this register.
How to run
Safetensors weights are on the Hugging Face repository. The parameter count on this page came from that index.
The Hugging Face model API lists the library as vllm.
Load it from huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602. This page does not add install steps that the Hub files do not state.
Agent capabilities
This model publishes no chat template, so there is nothing to inspect. That is not evidence of missing support — many base models and some releases simply ship without one.
Questions this record can answer
- When was Voxtral-Mini-4B-Realtime-2602 released?
- The Hugging Face repository was created on 21 Jan 2026, which is the date this register uses.
- How many parameters does Voxtral-Mini-4B-Realtime-2602 have?
- The safetensors index gives 4,429,679,360 parameters, shown here as 4.4B. That figure is exact, not rounded.
- What is the context length of Voxtral-Mini-4B-Realtime-2602?
- 1,500 tokens. That is max_position_embeddings from the model's own config.json — the architectural maximum, not necessarily the context a provider serves.
- What licence is Voxtral-Mini-4B-Realtime-2602 under?
- The Hub card lists apache-2.0.
- Is Voxtral-Mini-4B-Realtime-2602 gated?
- No. The Hugging Face model API does not mark this repository as gated.
Where each field came from
- Hugging Face model API — releasedAt, lastModified, downloads, likes, task, license, gated, parameters, libraryName, formats, baseModel. Checked 18 Oct 2026.
- config.json — contextLength, layers, vocabSize, architecture. Checked 18 Oct 2026.
First recorded 24 Sept 2026, last checked 18 Oct 2026. This record is compiled automatically from the sources above. Where a field is blank, we could not source it and have not guessed. Canonical repository: huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-2602.