# NVIDIA's Nemotron 3.5 Lightning Blends Mamba and MoE

> A 30B mixture-of-experts model with just 3B active parameters aims at fast, agentic coding workloads.

Published by The Open Weights on Aug 1, 2026. Canonical: https://theopenweights.com/news/nemotron-3-5-lightning-30b-a3b-ui67

## Key facts

- Company: NVIDIA
- Model: Nemotron-3.5-Lightning-30B-A3B
- Version: 3.5 Lightning
- Category: Text / LLM
- Modalities: Text / LLM, Reasoning
- License: Other (Open weights, commercial use allowed)
- Parameters: 30B (mixture of experts)
- Active parameters: 3B
- Context window: 262K tokens
- Architecture: NemotronHForCausalLM
- Precision: BF16
- Weights size: 65.8 GB
- Significance: notable
- Published: 2026-08-01
- Last verified: 2026-08-19
- Hugging Face: https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16
- Canonical URL: https://theopenweights.com/news/nemotron-3-5-lightning-30b-a3b-ui67

NVIDIA has published [Nemotron 3.5 Lightning 30B-A3B](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16) on Hugging Face, a text-generation and reasoning model that leans on an unusual architecture to keep inference costs down. It is a mixture-of-experts design with 30 billion total parameters but only about 3 billion active per token, released here in BF16 precision.

The standout detail is the hybrid layout: the model combines Mamba-style state-space components with a mixture-of-experts structure. That pairing is meant to deliver the throughput advantages of both approaches — Mamba's linear-scaling sequence handling and MoE's ability to route work to a small slice of the network at a time.

## Built for agentic coding

NVIDIA positions the release specifically for agentic coding, the kind of multi-step, tool-using workflows where a model plans, edits, and iterates rather than answering a single prompt. A few things stand out for that use case:

- Only 3B parameters activate per token, which lowers the compute and memory bill for long, repeated inference loops.
- The reasoning modality suggests it is tuned to produce structured, multi-step outputs.
- The compact active footprint makes it a candidate for teams that want capable coding assistance without frontier-scale serving costs.

Why it matters: agentic tools call models constantly, so per-token efficiency compounds fast. A sparse, hybrid model that keeps quality up while cutting active compute is exactly the trade-off developers building autonomous coding agents have been looking for.

The model ships under a custom NVIDIA license rather than a standard open license, so teams should review the terms before deploying it. Context length and benchmark figures were not specified in the release record.

## Get the model

- [Hugging Face](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16)

## Sources

- [nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16](https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16) — Hugging Face, Aug 1, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "NVIDIA's Nemotron 3.5 Lightning Blends Mamba and MoE", The Open Weights, Aug 1, 2026, https://theopenweights.com/news/nemotron-3-5-lightning-30b-a3b-ui67