Context window
256,000 tokens
Checked 19 Oct 2026https://docs.mistral.ai/models/mistral-small-4-0-26-03
LLM · Mistral
Mistral Small 4
Mistral's model page, dated 16 March 2026, labels Mistral Small 4 GA and Apache 2.0, with 119B parameters, 6.5B active, and a 256k context. The pricing page lists $0.15 input and $0.60 output per 1M tokens. The mistralai Hugging Face card posts the weights for Mistral-Small-4-119B-2603 and says the model is licensed under Apache 2.0.
Checked 19 Oct 2026https://docs.mistral.ai/models/mistral-small-4-0-26-03
Input price
$0.15
Checked 19 Oct 2026https://docs.mistral.ai/inference/pricing
Output price
$0.60
Checked 19 Oct 2026https://docs.mistral.ai/inference/pricing
API deployment
Yes
Checked 19 Oct 2026https://huggingface.co/mistralai/Mistral-Small-4-119B-2603
Local / self-hosted
Yes
Checked 19 Oct 2026https://huggingface.co/mistralai/Mistral-Small-4-119B-2603
VPC / cloud native
Not published
Checked 19 Oct 2026https://huggingface.co/mistralai/Mistral-Small-4-119B-2603
License
Apache 2.0
Checked 19 Oct 2026https://huggingface.co/mistralai/Mistral-Small-4-119B-2603
Data retention
Not published
Checked 19 Oct 2026https://docs.mistral.ai/models/mistral-small-4-0-26-03
Parameters
6.5B active / 119B total
Checked 19 Oct 2026https://docs.mistral.ai/models/mistral-small-4-0-26-03