Nemotron 3.5 Bolt 74B-A9B 24GB
The October 15 OpenMDW-1.1 release compresses a 74B-parameter multimodal agent model, with 9B active parameters, to a 24GB packed footprint using 2-bit quantization. It supports up to 256K context; runtime memory depends on configuration.
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Why this model scores 76.2
Published coding, browsing and tool-use results support meaningful local-agent capability, below the larger Super model. Small hardware requirements and released weights raise deployment pressure. Safety guidance is not a proof of containment, while lower capability moderates misuse relative to larger frontier models.
News tied to Nemotron 3.5 Bolt 74B-A9B 24GB
The model score of 76.2 rates this model's risk profile. The overall Doom Index of 69.1 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.
NVIDIA releases Nemotron 3.5 Super VL and a 24GB Bolt model for local agents
NVIDIA released open-weight Nemotron 3.5 Super VL and Bolt 74B-A9B 24GB. The models support multimodal agent workflows; Bolt compresses a larger model for local deployment, while Super VL supports up to one million context tokens.
- Full item contribution
- +0.07
- Nemotron 3.5 Bolt 74B-A9B 24GB equal share
- +0.04
Model score history
-
R1
Doom Score 76.2
Initial exact-version assessment from dated release and access evidence found during the owner-requested October re-audit.
30 Oct 2026