---
type: "post"
id: "nemotron-3-5-lightning-is-a-30b-hybrid-5ce211"
slug: "nemotron-3-5-lightning-is-a-30b-hybrid-5ce211"
title: "Nemotron 3.5 Lightning is a 30B hybrid Mamba-2 plus MoE plus Attention model with 3B active parameters."
description: "NVIDIA published NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, the BF16 reference weights for Nemotron 3.5 Lightning."
retrieval_nugget: "NVIDIA published NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, the BF16 reference weights for Nemotron 3.5 Lightning."
published_at: "2026-10-25"
updated_at: "2026-10-25"
record_date: "2026-10-25"
date_kind: "published_at"
topics: ["models","open-models"]
entities: ["NVIDIA","NVIDIA Nemotron 3.5 Lightning 30B-A3B-BF16","Nemotron","Mamba-2","OpenMDW 1.1","vLLM"]
related_patterns: []
novelty_label: "same_event"
what_changed: "Same event as NVIDIA Nemotron 3.5 Lightning."
prior_refs: ["memory:747251b9-1c05-8861-a655-a3531af4db77","memory:cd9342db-eb84-8398-ab13-6acc66fbff85"]
source_url: "https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16"
source_urls: ["https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:nemotron-3-5-lightning-is-a-30b-hybrid-5ce211"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/nemotron-3-5-lightning-is-a-30b-hybrid-5ce211/","markdown":"https://newruntime.com/posts/nemotron-3-5-lightning-is-a-30b-hybrid-5ce211.md","json":"https://newruntime.com/posts/nemotron-3-5-lightning-is-a-30b-hybrid-5ce211.json"}
---

# Nemotron 3.5 Lightning is a 30B hybrid Mamba-2 plus MoE plus Attention model with 3B active parameters.

## Retrieval answer

NVIDIA published NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, the BF16 reference weights for Nemotron 3.5 Lightning.

NVIDIA published NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16, the BF16 reference weights for Nemotron 3.5 Lightning. The model card lists an August 11, 2026 release date and says use is governed by the OpenMDW 1.1 model license.

The model has 30 billion total parameters and 3 billion active parameters per token. Its hybrid architecture combines Mamba-2, mixture-of-experts layers, and selected attention layers. NVIDIA positions the BF16 checkpoint for post-training, SFT, RL, distillation, domain adaptation, and creation of quantized or GGUF variants; it directs latency- and throughput-oriented deployment to a separate NVFP4 release.

The card lists single-GPU deployment on an 80GB H100 or A100. It validates 256K context on one H100, while 1M-token context is listed for GB200, B200, or an eight-H100 configuration. NVIDIA also documents configurable reasoning mode and a vLLM serving recipe.

NVIDIA reports benchmark results under its NeMo Gym and NeMo Evaluator harnesses and notes that results may differ from vendors' self-reported figures. The source does not establish independent benchmark performance, real-world cost, throughput, or performance on a particular workload.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- topic hub: [Models](https://newruntime.com/topics/models.md): Explore the models topic hub.
- topic hub: [Open Models](https://newruntime.com/topics/open-models.md): Explore the open-models topic hub.
- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)