Nemotron 3.5 Lightning (30B A3B) (Reasoning)

by NVIDIA

Nemotron 3.5 Lightning (30B A3B) (Reasoning) is a reasoning model from NVIDIA, released on August 11, 2026. It has a 262K-token context window and can generate up to 262K output tokens. API pricing is $0.08 per 1M input tokens and $0.2 per 1M output tokens. It ranks #114 of 220 models by Intelligence Score with a score of 21.9.

NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with ~3B active parameters, built for high-throughput specialized task execution in long-running / always-on agents. It ships with speculative decoding (MTP, DFlash, DSpark), NVFP4 and BF16 checkpoints, and is designed to pair with NVIDIA NeMo Switchyard for routing execution work alongside frontier planners. Released with open weights, data, and recipes under OpenMDW-1.1.

Key specs

ProviderNVIDIA
Release date2026-08-11
Licenseopen weights
Context window262K tokens
Max output262K tokens
Input price$0.08 / 1M tokens
Output price$0.2 / 1M tokens
Intelligence Score21.9
Coding index8.1

Quick answers

What is the context window of Nemotron 3.5 Lightning (30B A3B) (Reasoning)?

Nemotron 3.5 Lightning (30B A3B) (Reasoning) has a context window of 262K tokens, with a maximum output of 262K tokens per response.

How much does Nemotron 3.5 Lightning (30B A3B) (Reasoning) cost?

Nemotron 3.5 Lightning (30B A3B) (Reasoning) costs $0.08 per 1M input tokens and $0.2 per 1M output tokens through the API.

Is Nemotron 3.5 Lightning (30B A3B) (Reasoning) open source?

Yes. Nemotron 3.5 Lightning (30B A3B) (Reasoning) is an open model released under the open weights license.

When was Nemotron 3.5 Lightning (30B A3B) (Reasoning) released?

NVIDIA released Nemotron 3.5 Lightning (30B A3B) (Reasoning) on August 11, 2026.

How does Nemotron 3.5 Lightning (30B A3B) (Reasoning) compare to Claude Opus 5.5 (Reasoning)?

Nemotron 3.5 Lightning (30B A3B) (Reasoning) scores 21.9 on the Intelligence Score versus 60.5 for Claude Opus 5.5 (Reasoning), the current #1. Input pricing is $0.08 vs $4 per 1M tokens.

Compare on the AI leaderboard · All AI models

More models from NVIDIA