Skip to main content
AI Choice Engine

NVIDIA

NVIDIA Nemotron 3.5 Lightning 30B A3B

NVIDIA Nemotron 3.5 Lightning 30B A3B is a NVIDIA model with 1.05M context, Not verified in / Not verified out per million tokens, last verified October 2, 2026. Open weights: yes. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

NVIDIA's open-weight hybrid model for text generation and long-running agent workflows, distinct from Nemotron 3 Ultra.

Recorded profile verification October 2, 2026Individual provider fields may changeAccess: preview

AI model specification details
SpecificationNVIDIA Nemotron 3.5 Lightning 30B A3B
ProviderNVIDIA
TierBalanced
Context window1.05M
Max outputNot verifiedUnverified
Input / 1M tokensNot verifiedUnverified
Output / 1M tokensNot verifiedUnverified
WeightsOpen
Parameters30B total / 3B active (vendor rounded counts; Mamba-2/MoE/attention hybrid)
Reasoning levelsNot verifiedUnverified
Modalitiestext
LicenseOpenMDW License Agreement v1.1
API model idnvidia/nemotron-3.5-lightning-30b-a3b
ReleasedAugust 11, 2026

Pricing tiers: No standard direct per-token price or universal free quota is verified. Downloadable weights require compute; NVIDIA hosted trial and OpenCode Free-route access have separate terms and limits.

Best for

  • Self-hosted text agents
  • Long-context workflows
  • Sub-agent workloads

Watch out

The August 11 reference identifies version 1.0-preview. Native 1,048,576 context is from the exact NVFP4 checkpoint config; the hosted route advertises up to 1M without an exact route cap. Max output is unverified; sample max_tokens/reasoning_budget are examples. NVIDIA's hosted sample uses enable_thinking and reasoning_budget, not a confirmed effort enum. BF16 and NVFP4 have different vendor evaluation scores; these are not independent AA/SWE/Terminal leaderboard observations and are withheld from numeric comparisons pending exact configuration reconciliation. The OpenCode route's served precision is unverified.

History

NVIDIA Nemotron 3.5 Lightning 30B A3B timeline

Release and status events for this model that passed two-source verification — dated, sourced, and linked.

Source receipts

Catalog figures for this model were checked against the following sources.

Common questions

NVIDIA Nemotron 3.5 Lightning 30B A3B

Answered from the verified figures on this page rather than general guidance.

What is NVIDIA Nemotron 3.5 Lightning 30B A3B's context window?
NVIDIA Nemotron 3.5 Lightning 30B A3B accepts about 1.05M tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is NVIDIA Nemotron 3.5 Lightning 30B A3B best for?
NVIDIA Nemotron 3.5 Lightning 30B A3B is a balanced tier from NVIDIA. It suits self-hosted text agents, long-context workflows, sub-agent workloads. The August 11 reference identifies version 1.0-preview. Native 1,048,576 context is from the exact NVFP4 checkpoint config; the hosted route advertises up to 1M without an exact route cap. Max output is unverified; sample max_tokens/reasoning_budget are examples. NVIDIA's hosted sample uses enable_thinking and reasoning_budget, not a confirmed effort enum. BF16 and NVFP4 have different vendor evaluation scores; these are not independent AA/SWE/Terminal leaderboard observations and are withheld from numeric comparisons pending exact configuration reconciliation. The OpenCode route's served precision is unverified.
Can I self-host NVIDIA Nemotron 3.5 Lightning 30B A3B?
NVIDIA Nemotron 3.5 Lightning 30B A3B publishes open weights, but self-hosting depends on the licence, hardware footprint, quantisation quality, and serving stack. A hosted API is often cheaper until you have measured throughput and concurrency on your own hardware.
Model comparison catalogBenchmark library