Mistral Small 3.1
Pilot $4,900 fixed
- Sizes
- 24B dense
- Active params
- All parameters (dense)
- Context
- 128K
- Licence
- Apache 2.0
- Fits on
- 1× T4 16GB at 4-bit, 1× RTX 5090 32GB at FP8
- Endpoint
- llm.yourcompany.internal/v1
- Egress
- 0
- Owner
- YOU
Mistralon your own GPUs.
Mistral Small and Large on your own GPUs, behind an OpenAI-compatible endpoint, with no traffic leaving the network.
This card is the installation. Pick a size on its edge and it is re-issued.
The 72-hour pilot with Mistral.
$4,900 fixed. Small 3.1 runs on 1× T4 16GB at 4-bit. Small 4 is a 119B mixture-of-experts model that fits on 1× H100 80GB at 4-bit. Large 3 is the Apache 2.0 flagship and needs 8× H100 (640GB) at 4-bit.
Live on your data
72 h$4,900 fixed
Mistral live on your data in 72 hours. The clock starts after access and prerequisites.
Your endpoint
/v1OpenAI-compatible
OpenAI-compatible endpoint. Your applications switch by changing the base URL.
Your hardware
- Mistral Small 3.11× T4 16GB at 4-bit, 1× RTX 5090 32GB at FP8
- Mistral Small 41× H100 80GB at 4-bit, 2× H100 (160GB) at FP8
- Mistral Large 21× H100 80GB at 4-bit, 2× H100 (160GB) at FP8
- Mistral Large 38× H100 (640GB) at 4-bit, 8× H200 (1,128GB) at FP8
The readout
Evalon your data
Reference evaluation on your data. Readout and production scoping.
We size the hardware and source it at cost, with no markup. Ballpark: one H100 80GB ≈ $25–30k; a 4×H100 node ≈ $180–240k; smaller models run on far less.
Same SDK. Change the base URL.
The model is one parameter. Everything else in your application stays as it is.
client = OpenAI(
base_url="https://llm.yourcompany.internal/v1", # inside your network
api_key="your-key",
)
reply = client.chat.completions.create(
model="mistral-small-3-1",
messages=[{"role": "user", "content": "Summarise this contract."}],
)Licence, plainly.
- Mistral Small 3.1Apache 2.0Commercial use: permitted
- Mistral Small 4Apache 2.0Commercial use: permitted
- Mistral Large 2Mistral Research License (MRL) · research onlyCommercial use: not permitted
- Mistral Large 3Apache 2.0Commercial use: permitted
Mistral Small 3.1, Small 4 and Large 3 are Apache 2.0: commercial use, modification and private deployment are permitted.
Mistral Large 2 (123B) is not. It ships under the Mistral Research License, which does not allow commercial use without a separate paid licence from Mistral AI. For a commercial deployment of a Large model we use Large 3.
We read the current licence text against your use when we scope the pilot.
Published by Mistral AI. We are not affiliated with the vendor. This is a summary, not legal advice.
Related models.
- DevstralMistral AI1× T4 16GB at 4-bitApache 2.0
- Gemma 4Google2 models, from 1× T4 16GB at 4-bitApache 2.0
- Qwen3AlibabaCPU only at 4-bitApache 2.0
Talk to an engineer.
Tell us the model, the data and the GPUs you have. An engineer from Limassol or Auckland scopes the 72-hour pilot with you.