Introducing Mistral Small 4
Original: Mistral Small 4
Mistral Small 4 is an Apache 2.0 open model combining reasoning, multimodality, and agentic coding.
Mistral Small 4 is the next major release in the Mistral Small family, unifying Magistral-style reasoning, Pixtral-style multimodality, and Devstral-style coding agents. It uses a MoE architecture with 119B total parameters, 6B active parameters per token, a 256k context window, and configurable reasoning effort. The model is available via Mistral API, AI Studio, Hugging Face, open-source serving stacks, and NVIDIA deployment options.
Mistral AI has released Mistral Small 4, positioning it as the next major version of the Mistral Small series. The focus this time is not simply on improving chat-model capabilities, but on integrating several of Mistral's specialized capability lines into a single general-purpose model: Magistral's reasoning ability, Pixtral's multimodal input ability, and Devstral's agentic coding ability are now all placed into Mistral Small 4. For developers and enterprise teams, this means there is no longer a frequent need to switch between different specialized models when handling general chat, long-document understanding, image input, code automation, codebase exploration, and complex reasoning tasks.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Mistral AI News →Related
Summaries are AI-generated; the original article is authoritative.