ModelBench
NVIDIA logo
Models/NVIDIA

NVIDIA: Nemotron 3.5 Lightning 30B A3B

Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads

Source-linked nemotron Open weights Released 2026-08-11
API record Report
InputT
OutputT
Input price$0.05/M
Output price$0.2/M
Context262.144K
Max output262.144K
Providers8
Inference availability

Providers

Provider-specific identifiers, limits, and listed prices per million tokens. Every row links back to the provider's own documentation.

Report a price
Providers offering Nemotron 3.5 Lightning 30B A3B
ProviderProvider model IDContextMax outputInputOutputCache readCapabilitiesDocs
Nvidia nvidia/nemotron-3.5-lightning-30b-a3b 262.144K 262.144K — — — ReasoningToolsJSON Docs ↗
Vercel AI Gateway nvidia/nemotron-3.5-lightning 262.144K 131.072K — — — ReasoningToolsJSON Docs ↗
Merge Gateway nvidia/nemotron-3.5-lightning-30b-a3b 1M 262.144K — — — ReasoningToolsJSON Docs ↗
RunInfra nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16 262.144K 32.768K $0.05 $0.15 — ReasoningToolsJSON Docs ↗
Fireworks AI accounts/fireworks/models/nemotron-lightning-3p5-30b-a3b 262.144K 262.144K $0.05 $0.2 $0.01 ReasoningToolsJSON Docs ↗
NanoGPT nvidia/nemotron-3.5-lightning 1M 65.536K $0.05 $0.2 $0.01 ReasoningToolsJSON Docs ↗
Kilo Gateway nvidia/nemotron-3.5-lightning 262.144K 131.072K $0.08 $0.2 $0.04 ReasoningToolsJSON Docs ↗
OpenRouter nvidia/nemotron-3.5-lightning 1M 131.072K $0.08 $0.2 $0.04 ReasoningToolsJSON Docs ↗

Capability badges appear only where the provider catalog explicitly lists support. A blank cell means the source is silent, not that the feature is absent.

Listed rates

Price across providers

Input price per million tokens as published by each provider. Bars are drawn from listed rates only — no traffic weighting, since the catalog observes no requests.

Lowest input $0.05/M

Across 5 priced providers

Median input $0.05/M

Midpoint of listed rates

Highest input $0.08/M

1.6× the lowest listed rate

Output range $0.15 – $0.2

Per million output tokens

RunInfra $0.05/M
Fireworks AI $0.05/M
NanoGPT $0.05/M
Kilo Gateway $0.08/M
OpenRouter $0.08/M
Cost calculator

Estimate a workload

$0.00
Excludes taxes, non-token charges, and tiered discounts.

3 providers list the identical $0.05 input rate, so price alone will not separate them — compare context limits, max output, and capabilities above.

Context limits also differ by provider, from 262.144K to 1M tokens. Compare the provider table above before choosing on price alone.

Specification

Capabilities

Recorded from the source catalog and provider listings.

✓ Reasoning Yes
✓ Tool calling Yes
✓ Structured output Yes
× Attachments No
× Vision input No
✓ Open weights Yes
Creator
NVIDIA
Model family
nemotron
Knowledge cutoff
Not documented
License
Not documented
Release date
2026-08-11
Model ID
nvidia/nemotron-3.5-lightning
Published evaluations

Benchmarks

Every result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.

Benchmark registry
Benchmark profile
Coding IndexCoding Index Agentic IndexAgentic Index IntelligenceIntelligence Index This model — Coding Index: 26.8 index (31.0th percentile) This model — Agentic Index: 13.8 index (35.3th percentile) This model — Intelligence Index: 23.6 index (27.9th percentile)

Each axis is this model's percentile among the 3 benchmarks it has published results for, measured against every other model with a score on that same benchmark. Percentiles are used because benchmarks do not share a scale — a 60 on one is not a 60 on another. Hover any point for the raw score.

Coding Index index Source ↗
Coding Index #116 of 168
Agentic Index index Source ↗
Agentic Index #101 of 156
Intelligence Index index Source ↗
Intelligence Index #111 of 154

Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.

Cost against capability

Price and performance

Listed input price plotted against Coding Index, the benchmark with the widest published coverage that this model appears in.

0 41 83 $0.01 $0.1 $1 $10 GPT-5.6 Luna — $0.2/M, score 71.4 GPT-5.6 Terra — $2/M, score 76.7 GPT-5.6 Sol — $5/M, score 77.4 SpaceXAI: Grok 4.5 — $2/M, score 72.4 Claude Sonnet 5 — $2/M, score 71.5 Z.ai: GLM 5.2 — $0.476/M, score 68.8 Kimi K2.7 Code — $0.95/M, score 60.8 Claude Fable 5 — $10/M, score 76.5 Nex AGI: Nex-N2-Pro — $0.25/M, score 59.1 Nemotron 3 Ultra 550B A55B — $0.5/M, score 49.3 Qwen: Qwen3.7 Plus — $0.32/M, score 55.9 MiniMax: MiniMax M3 — $0.3/M, score 58.6 Step 3.7 Flash — $0.185/M, score 39.6 Anthropic: Claude Opus 4.8 — $5/M, score 74.3 Qwen: Qwen3.7 Max — $1.475/M, score 66.0 SpaceXAI: Grok Build 0.1 — $1/M, score 51.5 Gemini 3.5 Flash — $1.5/M, score 70.1 inclusionAI: Ring-2.6-1T — $0.075/M, score 42.8 SpaceXAI: Grok 4.3 — $1.25/M, score 42.2 IBM: Granite 4.1 8B — $0.05/M, score 9.5 Mistral: Mistral Medium 3.5 — $1.5/M, score 46.9 Qwen: Qwen3.6 35B A3B — $0.14/M, score 41.9 Qwen: Qwen3.6 27B — $0.289/M, score 53.7 GPT-5.5 — $5/M, score 74.9 DeepSeek V4 Pro — $0.435/M, score 58.7 DeepSeek V4 Flash — $0.14/M, score 56.2 Hy3 preview — $0.066/M, score 58.8 MiMo-V2.5-Pro — $0.435/M, score 60.2 MiMo-V2.5 — $0.14/M, score 56.8 inclusionAI: Ling-2.6-flash — $0.01/M, score 25.3 Kimi K2.6 — $0.95/M, score 61.8 Anthropic: Claude Opus 4.7 — $5/M, score 73.6 Z.ai: GLM 5.1 — $0.966/M, score 55.8 Gemma 4 26B A4B IT — $0.06/M, score 39.3 Gemma 4 31B IT — $0.08/M, score 43.4 Qwen: Qwen3.6 Plus — $0.325/M, score 54.5 Trinity Large Thinking — $0.22/M, score 25.8 Kwaipilot: KAT-Coder-Pro V2 — $0.3/M, score 59.5 MiniMax: MiniMax M2.7 — $0.3/M, score 52.6 GPT-5.4 nano — $0.2/M, score 56.1 GPT-5.4 mini — $0.75/M, score 56.1 Mistral: Mistral Small 4 — $0.15/M, score 26.6 Nemotron 3 Super 120B A12B — $0.2/M, score 37.7 Qwen: Qwen3.5-9B — $0.1/M, score 28.7 GPT-5.4 — $2.5/M, score 71.1 Inception: Mercury 2 — $0.25/M, score 31.1 Gemini 3.1 Flash Lite Preview — $0.25/M, score 34.7 Qwen: Qwen3.5-35B-A3B — $0.225/M, score 37.0 Qwen: Qwen3.5-122B-A10B — $0.29/M, score 45.7 Gemini 3.1 Pro Preview — $2/M, score 68.8 Anthropic: Claude Sonnet 4.6 — $3/M, score 63.0 Qwen: Qwen3.5 397B A17B — $0.39/M, score 48.2 Qwen: Qwen3 Coder Next — $0.12/M, score 36.2 Kimi K2.5 — $0.6/M, score 46.8 Upstage: Solar Pro 3 — $0.15/M, score 16.2 Z.ai: GLM 4.7 — $0.4/M, score 45.3 Nemotron 3 Nano 30B A3B — $0.05/M, score 14.4 Mistral: Devstral 2 2512 — $0.4/M, score 31.3 Amazon: Nova 2 Lite — $0.3/M, score 23.0 Mistral: Ministral 3 14B 2512 — $0.2/M, score 14.4 Mistral: Ministral 3 8B 2512 — $0.15/M, score 9.7 Mistral: Ministral 3 3B 2512 — $0.1/M, score 4.8 Mistral: Mistral Large 3 2512 — $0.5/M, score 20.1 DeepSeek V3.2 — $0.18/M, score 44.2 GPT-5.1 — $1.25/M, score 49.4 Kimi K2 Thinking — $0.6/M, score 21.0 Anthropic: Claude Haiku 4.5 — $1/M, score 43.9 Z.ai: GLM 4.6 — $0.5/M, score 45.8 Anthropic: Claude Sonnet 4.5 — $3/M, score 52.1 DeepSeek: DeepSeek V3.1 Terminus — $0.27/M, score 43.5 Qwen: Qwen3 Next 80B A3B Thinking — $0.15/M, score 17.4 Qwen: Qwen3 30B A3B Thinking 2507 — $0.2/M, score 12.1 Mistral: Mistral Medium 3.1 — $0.4/M, score 20.5 GPT-5 — $1.25/M, score 37.8 GPT-5 Mini — $0.25/M, score 15.6 GPT OSS 120B — $0.03/M, score 30.4 GPT OSS 20B — $0.03/M, score 20.7 Qwen: Qwen3 235B A22B Thinking 2507 — $0.23/M, score 22.1 Gemini 2.5 Pro — $1.25/M, score 33.3 Anthropic: Claude Sonnet 4 — $3/M, score 37.6 Google: Gemma 3n 4B — $0.06/M, score 3.2 Qwen: Qwen3 8B — $0.117/M, score 9.0 Qwen: Qwen3 14B — $0.12/M, score 13.8 Qwen: Qwen3 32B — $0.08/M, score 15.3 GPT-4.1 mini — $0.4/M, score 20.2 GPT-4.1 nano — $0.1/M, score 11.1 Meta: Llama 4 Maverick — $0.2/M, score 16.3 Meta: Llama 4 Scout — $0.1/M, score 8.2 DeepSeek: DeepSeek V3 0324 — $0.27/M, score 21.2 Google: Gemma 3 4B — $0.05/M, score 2.7 Google: Gemma 3 12B — $0.05/M, score 5.8 Cohere: Command A — $2.5/M, score 27.8 Google: Gemma 3 27B — $0.08/M, score 10.1 OpenAI: o3 Mini High — $1.1/M, score 16.3 DeepSeek-R1 — $0.7/M, score 24.6 o1 — $15/M, score 39.7 Meta: Llama 3.3 70B Instruct — $0.1/M, score 11.9 Meta: Llama 3.1 8B Instruct — $0.05/M, score 5.4 GPT-4o mini — $0.15/M, score 11.4 GPT-4o (2024-05-13) — $5/M, score 24.2 GPT-4 Turbo — $10/M, score 21.5 GPT-4 — $30/M, score 13.1 GPT-3.5-turbo — $0.5/M, score 10.7 Muse Spark 1.1 — $1.25/M, score 71.3 Kimi K3 — $3/M, score 76.2 Inkling — $1.87/M, score 52.1 Gemini 3.6 Flash — $1.5/M, score 69.2 Gemini 3.5 Flash Lite — $0.3/M, score 49.3 Claude Opus 5 — $5/M, score 78.0 Google: Gemini 3.6 Flash (batch) — $0.375/M, score 69.2 Google: Gemini 3.5 Flash Lite (batch) — $0.15/M, score 49.3 Anthropic: Claude Sonnet 5 (batch) — $1/M, score 71.5 Anthropic: Claude Fable 5 (batch) — $5/M, score 76.5 MiniMax: MiniMax M3 (batch) — $0.15/M, score 58.6 Anthropic: Claude Opus 4.8 (batch) — $2.5/M, score 74.3 Google: Gemini 3.5 Flash (batch) — $0.75/M, score 70.1 OpenAI: GPT-5.5 (batch) — $2.5/M, score 74.9 Anthropic: Claude Opus 4.7 (batch) — $2.5/M, score 73.6 OpenAI: GPT-5.4 Nano (batch) — $0.1/M, score 56.1 OpenAI: GPT-5.4 Mini (batch) — $0.375/M, score 56.1 OpenAI: GPT-5.4 (batch) — $1.25/M, score 71.1 Google: Gemini 3.1 Pro Preview (batch) — $1/M, score 68.8 OpenAI: GPT-5.1 (batch) — $0.625/M, score 49.4 Anthropic: Claude Haiku 4.5 (batch) — $0.5/M, score 43.9 Anthropic: Claude Sonnet 4.5 (batch) — $1.5/M, score 52.1 OpenAI: GPT-5 (batch) — $0.625/M, score 37.8 OpenAI: GPT-5 Mini (batch) — $0.125/M, score 15.6 Google: Gemini 2.5 Pro (batch) — $0.625/M, score 33.3 DeepSeek V4 Flash 0731 — $0.08/M, score 69.1 Inkling Small — $0.45/M, score 52.9 DeepSeek V4 Pro 0813 — $0.435/M, score 68.8 Qwen: Qwen3.8 Max — $2/M, score 71.8 Ling-3.0-flash — $0.021/M, score 50.6 Claude Opus 5 (batch) — $2.5/M, score 78.0 Thinking Machines: Inkling (batch) — $0.5/M, score 52.1 OpenAI: GPT-5.6 Luna (batch) — $0.1/M, score 71.4 OpenAI: GPT-5.6 Terra (batch) — $1/M, score 76.7 OpenAI: GPT-5.6 Sol (batch) — $1.25/M, score 77.4 Z.ai: GLM 5.2 (batch) — $0.7/M, score 68.8 MoonshotAI: Kimi K2.7 Code (batch) — $0.475/M, score 60.8 NVIDIA: Nemotron 3 Ultra (batch) — $0.3/M, score 49.3 Anthropic: Claude Sonnet 4.6 (batch) — $1.5/M, score 63.0 OpenAI: GPT-4.1 Mini (batch) — $0.2/M, score 20.2 OpenAI: GPT-4.1 Nano (batch) — $0.05/M, score 11.1 OpenAI: o3 Mini High (batch) — $0.55/M, score 16.3 OpenAI: o1 (batch) — $7.5/M, score 39.7 OpenAI: GPT-4o-mini (batch) — $0.075/M, score 11.4 OpenAI: GPT-4 Turbo (batch) — $5/M, score 21.5 OpenAI: GPT-3.5 Turbo (batch) — $0.25/M, score 10.7 Solar Pro 4 — $0.3/M, score 52.7 Nemotron 3.5 Lightning 30B A3B — $0.05/M, score 26.8 SpaceXAI: Grok 4.6 — $2/M, score 76.8 Gemini 3.7 Flash — $0.75/M, score 76.1 Google: Gemini 3.7 Flash (batch) — $0.188/M, score 76.1 Nemotron 3.5 Lightning 30B A3B Input price per million tokens (log scale) Index

The stepped line is the efficient frontier: at each price, the best score available for that money or less. A model sitting on it is not being beaten by anything cheaper. Price is log-scaled because listed rates span four orders of magnitude. Only models with both a listed price and a score on this benchmark can appear.

Catalog activity

Change log

Field-level changes detected between successful source imports.

Full change log
Max Output Tokens131072 → 262144
Price Completion0.19999999999999998 → 0.2
Price Prompt0.08 → 0.05
Max Output Tokens262144 → 131072
Price Completion0.2 → 0.19999999999999998
Price Prompt0.05 → 0.08
Max Output Tokens131072 → 262144
Price Completion0.19999999999999998 → 0.2
Price Prompt0.08 → 0.05
Max Output Tokens262144 → 131072
Provenance

Sources & verification

Every figure on this page traces back to one of these records.

Methodology
Last verified
Aug 17, 2026
Status
Source-linked
Confidence
Medium
Catalog source
Models.dev
Models.devcatalog ↗Nvidia documentationprovider documentation ↗Vercel AI Gateway documentationprovider documentation ↗Merge Gateway documentationprovider documentation ↗RunInfra documentationprovider documentation ↗Fireworks AI documentationprovider documentation ↗NanoGPT documentationprovider documentation ↗Kilo Gateway documentationprovider documentation ↗OpenRouter documentationprovider documentation ↗Intelligence Indexindependent benchmark ↗
Public API

Use this record

Fetch the complete source-linked model record. No key, no account, no rate-limited tier.

API documentation
Endpoint
GET https://www.modelbench.lol/api/v1/models/nvidia/nemotron-3.5-lightning
curl
curl "https://www.modelbench.lol/api/v1/models/nvidia/nemotron-3.5-lightning"
Common questions

Frequently asked questions

Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.

What is Nemotron 3.5 Lightning 30B A3B?

Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads. It is published by NVIDIA and catalogued here from Models.dev.

How much does Nemotron 3.5 Lightning 30B A3B cost?

Listed input pricing starts at $0.05 per million tokens from RunInfra, rising to $0.08 across 5 listed providers.

What is the context length of Nemotron 3.5 Lightning 30B A3B?

Nemotron 3.5 Lightning 30B A3B accepts up to 262.144K tokens of context and returns up to 262.144K output tokens.

Does Nemotron 3.5 Lightning 30B A3B support tool calling and structured output?

Provider catalogs list support for tool calling, structured output, and reasoning.

Which providers serve Nemotron 3.5 Lightning 30B A3B?

8 providers list this model: Nvidia, Vercel AI Gateway, Merge Gateway, RunInfra, Fireworks AI, NanoGPT and 2 more.

Are the weights for Nemotron 3.5 Lightning 30B A3B open?

Yes. The weights are published.

When was Nemotron 3.5 Lightning 30B A3B released?

The catalog records a release date of 2026-08-11, last verified Aug 17, 2026.

More models from NVIDIA

View all →
NVIDIA: Nemotron 3.5 Content Safety (free)

NVIDIA Nemotron 3.5 Content Safety is a compact 4B-parameter multimodal guardrail model from NVIDIA, fine-tuned from Google Gemma-3-4B. It moderates both inputs to and responses from LLMs and VLMs, accepting...

128K ctx NVIDIA: Nemotron 3 Ultra (free)

NVIDIA Nemotron 3 Ultra is an open frontier-reasoning and orchestration model from NVIDIA, with 55B active parameters out of 550B total (MoE). Built on a hybrid Transformer-Mamba mixture-of-experts architecture, it...

1M ctx Nemotron 3 Ultra 550B A55B

Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy

1M ctx $0.5/M in NVIDIA: Nemotron 3 Nano Omni (free)

NVIDIA Nemotron™ 3 Nano Omni is a 30B-A3B open multimodal model designed to function as a perception and context sub-agent in enterprise agent systems. It accepts text, image, video, and...

256K ctx