BlogLLM Benchmark Scores as Published, Side by Side

Devstral 2: Benchmark Scores as Published

Every result the catalog holds for Devstral 2: 1 test in 1 area, from 2 pages. It was printed by Mistral AI itself. Each number is as printed; nothing is ranked or rescored.

Developer Mistral AIReleased 2025-12-09Context window 256KWeights Open (Modified MIT License)Data version 2026103001, as of 2026-10-30

Agentic Coding

TestValuePrinted bySetup as printed
SWE-bench Verified72.2%Mistral AI, Devstral 2 and Mistral Vibe CLIchecked 2026-10-15; printed on 1 more page-

More from Mistral AI

Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.