Mistral Small

Mistral Small is an efficient model from Mistral AI for fast chat, extraction, and production assistants. It is part of the mistral-small line, succeeding Mistral Small 3.2. It has a context window of 256000 tokens and a max output of 256000 tokens. Weights are open. It supports reasoning mode, tool calling, and accepts image and text inputs, returning text. Pricing starts at $0.075 per million input tokens and $0.20 per million output tokens, with cached input at $0.015 per million tokens. Knowledge cutoff is 2025-06.

Against Mistral Small 3.2: window 128K to 256K, input $0.10 to $0.075, output $0.30 to $0.20.

Range across the hosts that serve it: up to $0.15 per 1M input tokens.

• ·256k token context and output window
• ·Open weights for self-hosting
• ·Supports reasoning and tool calling
• ·Accepts image and text inputs
• ·Priced from $0.075 per million input tokens

• Reach for it for fast chat assistants
• Reach for it for extraction from text and images
• Reach for it for production assistants with tool use
• Reach for it for long-context reasoning tasks

6 hosts serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window size?

The context window is 256000 tokens, and the max output is also 256000 tokens.

Does it support image input?

Yes, it accepts image and text inputs, and returns text output.

What is the pricing?

Input starts at $0.075 per million tokens, output at $0.20 per million tokens, and cached input at $0.015 per million tokens.