The AI Tokens

Mistral AI Models - Token Counter & Cost Calculator

Calculate token usage and API costs for Mistral's open-source language models. Get accurate pricing estimates for Mistral Large, Medium, Small, and Codestral.

Available Mistral Models

Mistral Large 2

Context: 128,000 tokens

Input: $2/1MOutput: $6/1M

Optimized for high reasoning and coding capabilities, competitive with GPT-4o but more cost-effective.

Mistral Medium

Context: 131,072 tokens

Input: $0.4/1MOutput: $2/1M

High-performance, cost-effective model designed for efficient, high-accuracy, and fast inference. Max output: 8,191 tokens.

Codestral 2

Context: 128,000 tokens

Input: $0.3/1MOutput: $0.9/1M

Optimized for coding, debugging, and software development, supporting large codebases.

Devstral 2

Context: 128,000 tokens

Input: $0.4/1MOutput: $2/1M

Released December 2025, specialized development model with enhanced coding capabilities.