Speech-to-text
Voxtral Mini 4B Realtime (2602) requirements, license & speed
By Mistral AI · released 2026-01-21 · 4.43B parameters
Commercial useYes
Min memorySingle GPU with >= 16 GB memory (BF16, vLLM)
Runs on CPUUnknown
StreamingYes
Key facts
- License
- apache-2.0 source
- Commercial use
- Yes sourceApache-2.0, card states research and commercial use
- Parameters
- 4.43B sourcesafetensors parameter count
- Minimum memory
- Single GPU with >= 16 GB memory (BF16, vLLM) source
- Runs on CPU
- Unknown
- Streaming
- Yes sourceNatively streaming, causal audio encoder
- Voice cloning
- n/a (speech-to-text)
- Languages
- en, fr, es, de, ru, zh, ja, it, pt, nl, ar, hi, ko source
- Released
- 2026-01-21 sourceHugging Face repo creation date (proxy for release date)
- Gated on HF
- No
Measured speed & memory
Benchmark pending. We haven't measured Voxtral Mini 4B Realtime (2602) yet. Planned targets: Raspberry Pi 5, CPU-only x86, Apple Silicon Mac and consumer GPUs. Get notified.
Accuracy
We don't run accuracy evals. For word error rate, check the Hugging Face Open ASR Leaderboard.
Similar STT models
Spotted an error? Report it. Every fact links to the model card or repo it came from.