Unverified50% confidenceEventExact time
The developer built an inference engine based on PyTorch and Triton and achieved Nemotron-3.5-Super running on an RTX 5060 (8GB VRAM).
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/17/2026
First Seen
Valid until: 1/15/2027
Sources
Related Entities
Related Claims
UnverifiedNemotron 3.5 Bolt controls reasoning modes via chat-template, using enable_thinking=True/False to toggle thinking mode, and the recommended sampling parameters are temperature 1.0 and top_p 0.95, both inherited from Super 3.5 VL.67% similarUnverifiedNemotron 3.5 Super VL's open-source deliverables include checkpoints, training recipes, and datasets, hosted on Hugging Face, with NIM containers also provided for deployment.66% similarUnverifiedNemotron 3.5 Super VL's backbone was initialized from Nemotron 3 Super Base, distilled 1T tokens via Nemotron 3 Ultra Base, extended to a 1M context, then underwent 67B tokens of software engineering mid-training, and only then was vision training done.66% similarUnverifiedNemotron 3.5 Super VL has 120 billion total parameters and 12 billion active parameters.65% similarUnverifiedThe larger-scale Nemotron Ultra version can be used directly on the Perplexity platform.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1053037API
curl https://kongchang.com/api/v1/knowledge/claims/1053037MCP
get_claim(id=1053037)