Skip to main content

Nemotron 3 Nano 30B A3B

NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully open with open-weights, datasets and recipes so developers can easily customize, optimize, and deploy the model on their infrastructure for maximum privacy and security.

Input / output price
$0.05 / $0.20 per 1M
Context
262K tokens
Released
Dec 15, 2025
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "nemotron-3-nano-30b-a3b".

Read docs

Providers

5 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Concentrateconcentrate/nemotron-3-nano-30b
$0.05/M$0.20/M
NovitaRetiringnovita/nemotron-3-nano-30b-a3b
$0.05/M$0.20/M
Bedrockbedrock/nemotron-3-nano-30b-a3b
$0.06/M$0.24/M
Nebiusnebius/nemotron-3-nano-30b-a3b
$0.06/M$0.24/M
OpenRouteropenrouter/nemotron-3-nano-30b-a3b
$0.06/M$0.24/M

Is a provider down? AWS Bedrock status, OpenRouter status

Pricing

Averaged across 5 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.06 per 1M
Average output
$0.22 per 1M
Average cache read
$0.02 per 1M

Cost by volume

Total cost from 0 to 1T tokens on the cheapest and most expensive provider.

Highest$0.06/MBedrock, Nebius, and OpenRouterLowest$0.05/MConcentrate and Novita

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

24tokens/sacross providers

Faster than 52% of the other 596 models on the gateway

Response latency

Lower is better

0.4sacross providers

Faster than 83% of the other 768 models on the gateway

Benchmarks

The Respan Index plus 10 published scores from 2 sources, each linked to where it was published.

15.6/ 100

Respan Index#158 of 176

176 ranked models. The shaded band is this model's 95% interval, 10.4 to 29.4. Amazon Nova 2 Lite and GPT-4.1 mini fall inside it, so the gap is within the margin of error.
Nemotron 3 Nano 30B A3B by category: Coding 3.5 (estimated), Agents 13.1 (estimated), Reasoning 21.0, Knowledge & language 10.4, Human preference 34.3
  1. 158NVIDIA Nemotron 3 Nano 30B-A3B BF1615.6
  2. 157Amazon Nova 2 Lite15.8
  3. 159GPT-4.1 mini15.3

Published scores

BenchmarkScore
LiveCodeBenchhuggingface.co68.3%
SWE-Bench (OpenHands)huggingface.co38.8%

Compare Nemotron 3 Nano 30B A3B: NVIDIA Nemotron 3 Nano 30B-A3B BF16 vs NVIDIA Nemotron 3.5 Super VL 120B-A12B NVFP4, NVIDIA Nemotron 3 Nano 30B-A3B BF16 vs NVIDIA Nemotron 3 Ultra 550B-A55B NVFP4, NVIDIA Nemotron 3 Nano 30B-A3B BF16 vs NVIDIA Nemotron 3 Super 120B-A12B, NVIDIA Nemotron 3 Nano 30B-A3B BF16 vs NVIDIA Nemotron 3.5 Lightning 30B-A3B NVFP4

Apps

Use Nemotron 3 Nano 30B A3B in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Nemotron 3 Nano 30B A3B FAQ

Short answers, built from the data on this page.

Nemotron 3 Nano 30B A3B is a model from NVIDIA, released on Dec 15, 2025. NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully open with open-weights, datasets and recipes so developers can easily customize, optimize, and deploy the model on their infrastructure for maximum privacy and security.

On the Respan gateway, Nemotron 3 Nano 30B A3B costs from $0.05 per 1M input tokens and $0.20 per 1M output tokens on Concentrate and Novita, after gateway discounts. Averaged across its 5 providers, it costs $0.06 per 1M input tokens and $0.22 per 1M output tokens.

Nemotron 3 Nano 30B A3B is available through 5 providers: Concentrate, Novita, Bedrock, Nebius, and OpenRouter.

Nemotron 3 Nano 30B A3B accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "nemotron-3-nano-30b-a3b". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Nemotron 3 Nano 30B A3B scores 15.6 out of 100 on the Respan Index v1.1, ranking #158 of 176 models. By category: Coding 3.5 (estimated), Agents 13.1 (estimated), Reasoning 21.0, Knowledge & language 10.4, and Human preference 34.3. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Nemotron 3 Nano 30B A3B has 10 published benchmark scores, including LiveCodeBench 68.3%, GPQA (with tools) 75%, and AIME 2025 (with tools) 99.2%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.