Magistral Small 2506
READER SUMMARYMistral's first small reasoning model, released as open weights with 40,000-token context and agent and tool interfaces.
Why this model scores 68.3
Explicit reasoning and tool interfaces support moderate agent capability. Open weights and broad local deployment sharply reduce provider control despite sub-frontier capability.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
News tied to Magistral Small 2506
The model score of 68.3 rates this model's risk profile. The overall Doom Index of 63.5 measures the recalibrated complete evidence corpus. These values answer different questions.
Each article's leave-one-out Doom Index contribution is divided equally among the exact models named on that article. This prevents multi-model evidence from being claimed in full on several model pages. Model risk scores are recalculated from their technical profile and related news, but never feed back into the overall index.
Mistral releases open and hosted Magistral reasoning models
Mistral released Magistral Small 2506 as open weights and Magistral Medium 2506 through its API and Le Chat, adding explicit reasoning to its model portfolio.
- Full item contribution
- +0.02
- Magistral Small 2506 equal share
- +0.01
Model score history
- R1Doom Score 68.3
Initial exact profile for Magistral Small 1.0 version 25.06.
12 Aug 2026