LLM
GA
Open weights
Voxtral
Local audio transcription, summarization, and voice assistant
Release2025-07
Context33K
Max output—
Parameters24B
Local / VRAM16–24 GB
LicenseOpen weights
Overview
Voxtral is a model in the LLM category from Mistral AI. It was released in 2025-07. The vendor is based in France. Best for: Local audio transcription, summarization, and voice assistant.
Strengths
- Audio input
- Apache-2.0
- 24B
Limits
- 32K context
- No vision
Specifications
| Vendor | Mistral AI |
|---|---|
| Category | LLM |
| Release | 2025-07 |
| Status | GA |
| License | Open weights |
| License name | Apache-2.0 |
| Commercial use | Allowed |
| Parameters | 24B |
| Architecture | audio-language model |
| Context window | 33K |
| Input modalities | Audio, Text |
| Output modalities | Text |
| Prompt types | Text |
| Negative prompts | No |
| Max outputs | 1 |
| Local deployment | Yes |
| VRAM | 16–24 GB |
| System RAM | 32–64 GB |
| API access | Yes |
| Pricing | Free weights |
| Confidence | High |
| Vendor country | France |
| Vendor type | Frontier lab |
Pricing & hardware
Free weights
This model can be self-hosted or run locally. Approximate GPU memory: 16–24 GB. Approximate system RAM: 32–64 GB. API access is available.