Unverified50% confidenceFactExact time
Nemotron 3.5 Bolt E4B 4GB is a derivative model obtained through post-training on top of Gemma 4 E4B, not a new architecture trained from scratch.
1
Sources
50%
Confidence
Long-term
Relevance
10/24/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedNVIDIA open-sourced a small language model called Nemotron 3.5 Bolt E4B 4GB.79% similarUnverifiedNVIDIA fine-tuned Nemotron 3.5 Bolt E4B based on Gemma 4 E4B, surpassing the original Gemma 4 E4B on 16 of 21 NVIDIA tests.79% similarUnverifiedCurrently available public information does not disclose the data scale, fine-tuning method details or benchmark results used in the post-training of Nemotron 3.5 Bolt E4B 4GB.77% similarUnverifiedThe problem of DPO training previously crashing for Gemma 4 E2B and E4B (including Nemotron versions) has been fixed.74% similarUnverifiedNemotron 3.5 Bolt 74B-A9B uses native 2-bit quantization; the model was designed for low-bit precision from the training stage rather than compressed after the fact.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1092495API
curl https://kongchang.com/api/v1/knowledge/claims/1092495MCP
get_claim(id=1092495)