Deploy Voxtral-4B-TTS-2603
Voxtral-4B-TTS-2603 by Mistral AI on GPU capacity reserved for you: a private OpenAI-compatible endpoint, every request processed on European GPUs and logged in your audit trail, billed per GPU-hour on a monthly commitment.
- Maker
- Mistral AI
- Params
- —
- Context
- —
- Architecture
- Dense
- License
- CC BY-NC 4.0
- Released
- 2025-11-17
Footprint
Minimum to load the published checkpoint with a small KV cache. Long contexts and high concurrency need more; quantizing a 16-bit checkpoint needs less. You don't pick hardware on sference: you choose a performance tier and we size and run the deployment.
License
CC BY-NC 4.0Non-commercial only
Non-commercial use only. Production use needs a separate commercial license from the model maker. Read the license. Not legal advice.
FAQ
Can I run Voxtral-4B-TTS-2603 on a dedicated deployment in Europe?
Yes. sference deploys Voxtral-4B-TTS-2603 on GPU capacity reserved for you, behind a private endpoint, with every request processed on European GPUs and logged in your audit trail. Tell us your expected volume and latency targets with the form on this page and we come back with a sizing and a quote.
How many GPUs does Voxtral-4B-TTS-2603 need?
The published checkpoint is about 8.0 GB (—), so loading it takes at least 1 GPU with 80 GB of memory each, or 1 GPU with 141 GB each. Long contexts and high concurrency need more for the KV cache; quantization needs less. You don't pick hardware on sference: you choose a performance tier and we size the deployment.
Is Voxtral-4B-TTS-2603 licensed for commercial use?
Voxtral-4B-TTS-2603 is published under the CC BY-NC 4.0 license. Non-commercial use only. Production use needs a separate commercial license from the model maker.
How is a dedicated deployment priced?
Per GPU-hour on a monthly commitment, sized to your throughput and latency targets.
Related models
Request Voxtral-4B-TTS-2603
Tell us roughly how much traffic you expect. A fine-tune of Voxtral-4B-TTS-2603? Mention it in the notes.
- 01We size it
A performance tier, the capacity behind it, and a quote for your volume.
- 02We benchmark it
Bring one workload with a sample of real traffic; we come back with latency and throughput numbers.
- 03It goes live
We provision the deployment. You call it with the same API and model id as before.
Specs from Hugging Face, synced 2026-10-27. Machine-readable version: https://sference.com/models/mistralai/Voxtral-4B-TTS-2603.md