Devstral

Mistral

Official devstral-2512 Mistral AI model

Capabilities

Tool Use

Technical Specifications

Context Window

262,144 tokens

Max Output

262,144 tokens

Pricing

Token Costs (per 1M tokens)

Cache Miss Input

$0.40

Non-Reasoning Output

$2

Tool Costs (per 1K calls)

Web Search

$15

Code Execution

$0.19

Legacy

Made legacy on February 20, 2026

Reason

Superseded by Devstral Medium with stronger agentic coding

Recommended Replacement