Atlas

Models

← All models

Devstral 2

Mistral AI · 2025-12-09 · 125B parameters

Devstral 2 is Mistral AI's dense, text-only, 123B-class open-weight agentic coding model, released as the larger member of the Devstral 2 family alongside Devstral Small 2. Its instruction-tuned FP8 checkpoint supports 262,144 tokens through YaRN extrapolation from an original 4,096-position setting and is designed for tool-driven codebase exploration, multi-file editing, and software-engineering agents.

Benchmark scores

BenchmarkScore
AA-Briefcase Elo483
AA-Briefcase v1.1 Elo483
AA-LCR30.0
AA-LCR v1.132.3
AA-Omniscience Index-46.7
AIME 2025 (Artificial Analysis avg@10)36.7
Artificial Analysis Agentic Index10.6
Artificial Analysis Coding Index31.3
Artificial Analysis Intelligence Index19.2
Artificial Analysis Intelligence Index v4.39.4
Artificial Analysis Intelligence Index v4.3.28.6
Artificial Analysis Omniscience Accuracy20.8
Artificial Analysis Omniscience Hallucination Rate85.3
Artificial Analysis Openness Index27.8
Artificial Analysis Progress Index v1197
AutomationBench-AA 1.0.63.1
BuseyBench SVG2.4
Capability136.8
CritPt (avg@5)0.0
GDP.pdf (Artificial Analysis, All-pass)2.4
GDP.pdf (Artificial Analysis, Mean Criteria)45.6
GDPval-AA v2689
GDPval-AA v2 (normalized)12.3
GDPval-AA v2.1552
Global MMLU Lite77.5
GPQA Diamond (Artificial Analysis avg@5)59.4
Humanity's Last Exam (Text-Only)3.6
IFBench (Artificial Analysis avg@5)38.1
IOI0.7
IOI - IOI 20240.0
Loading Atlas data…