DeepSeek-V3.2
MIT-licensed 685B hybrid reasoning model with thinking-with-tools, agentic post-training, downloadable weights, and hosted API access.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why this model scores 82.7
Frontier-level reasoning, tool use, and unrestricted downloadable weights produce high diffusion and residual control difficulty.
News tied to DeepSeek-V3.2
The model score of 82.7 rates this model's risk profile. The overall Doom Index of 67.9 measures the complete temporally weighted evidence record. These values answer different questions.
Each article's current 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 use a bounded temporal offset around their technical profile, but never feed back into the overall index.
DeepSeek-V3.2 open weights bring reasoning into agentic tool use
DeepSeek released MIT-licensed weights for DeepSeek-V3.2 and the higher-compute DeepSeek-V3.2-Speciale while upgrading its hosted chat and reasoning APIs to V3.2.
- Full item contribution
- +0.19
- DeepSeek-V3.2 equal share
- +0.10
Model score history
-
R3
Doom Score 82.7
Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-142416 under temporal-monthly-pressure-v4.
12 Aug 2026 -
R2
Doom Score 82.4
Exact-version evidence chronology replayed after run doombench-hourly-news-20260812-132112 under fixed-sensitivity-v3.
12 Aug 2026 -
R1
Doom Score 81.6
New exact model profile from dated API changelog and official model card.
11 Aug 2026