Devstral 2

Mistral AI llm

Mistral's flagship open-weight agentic coding model, released alongside the Mistral Vibe CLI.

overall ⌄

46.2

46 of 93 ranked

price · $/M tokens

0.4 / 2

in / out, as of 2026-07-25

to run it yourself ⌄

74 GB

minimum · 295 GB recommended

Benchmarks

What its maker published, and what anyone else measured. Every figure links the document it came from.

benchmark	published	measured
SWE-bench	72.2	—

On a 1,000-in, 1,000-out call these cost less and rank no lower — a hard comparison to argue with, and a narrow one. Value covers what it misses.

Ministral 3 14B Reasoning Mistral AI	$0.00040	overall 48.1
DeepSeek V4-Flash DeepSeek	$0.00042	overall 59.3
DeepSeek V4-Pro DeepSeek	$0.00130	overall 68.3
MiniMax M2.1 MiniMax	$0.00150	overall 50.9
MiniMax M3 MiniMax	$0.00150	overall 65.1

No recorded lineage — root or standalone model.

org	Mistral AI
released	2025-12-09
params	123B
license	modified-mit
hf id	mistralai/Devstral-2-123B-Instruct-2512