LLMOpen weights

Devstral 2

Mistral AI

Updated May 21, 2026 · Benchmarks via Artificial Analysis, specs & pricing via OpenRouter · Methodology · Report an error

Specifications

Type: LLM
Access: Open weights
Released: December 9, 2025
Context window: 262,144 tokens
Output speed: 51 tok/s
Latency (TTFT): 0.64s
Input: Text
Output: Text
API pricing: $0.40 in · $2.00 out / 1M tokens

Capabilities

Function callingStructured output

Benchmarks

Reasoning

59.4

Coding

44.8

Math

36.7

General

MMLU-Pro	76.2
Humanity’s Last Exam	3.6

Compare Devstral 2 with

vs Sonar Reasoning Pro vs R1 1776 vs Qwen3.7 Max vs Gemini 3.5 Flash

See all Devstral 2 alternatives →