# NVIDIA Releases Nemotron 3.5 Lightning With 1M-Token Context on One GPU — retained record

> Retained news record from news:ml-alphasignal.

Canonical: https://highsignal.app/data/records/37fab97565a0f294

[← sources](https://highsignal.app/data)


source: [news](https://highsignal.app/data/news) (news:ml-alphasignal)  published 2026-10-25 16:03 ingested 2026-10-26 02:56

[original source ↗](https://alphasignal.ai/news/nvidia-releases-nemotron-3-5-lightning-with-1m-token-context-on-one-gpu)

## retained text



```
NVIDIA's new 30B hybrid Mamba-Transformer MoE activates just 3B parameters, runs on a single H100, and handles up to 1M tokens of context.
```



document fetched 2026-10-26 02:56