Back to Models

Nemotron Nano 30B

nvidianvidia/nemotron-nano-30b

API guide

Chat completion

Standard OpenAI-compatible chat completion β€” the simplest way to call Nemotron Nano via TheRouter.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-nano-30b",
    "messages": [
      {"role": "user", "content": "What are the key design goals of the Nemotron 3 architecture?"}
    ]
  }'

Streaming

Stream tokens as they are generated β€” ideal for responsive chat UX.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-nano-30b",
    "messages": [{"role": "user", "content": "Explain hybrid Mamba-Transformer in one sentence."}],
    "stream": true
  }'

Tool calling

Nemotron Nano was specifically trained for agentic tool-use across multi-step environments via NeMo Gym RL.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-nano-30b",
    "messages": [{"role": "user", "content": "What is the weather in San Francisco?"}],
    "tools": [{
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get weather for a location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {"type": "string"}
          },
          "required": ["location"]
        }
      }
    }],
    "tool_choice": "auto"
  }'

JSON mode

Structured JSON output via response_format β€” useful for extracting structured data.

cURL
curl https://api.therouter.ai/v1/chat/completions \
  -H "Authorization: Bearer $THEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/nemotron-nano-30b",
    "messages": [{"role": "user", "content": "List three key features as JSON."}],
    "response_format": {"type": "json_object"}
  }'
Fact ledger β€” every claim on this page traces here
sourceURLretrieved
Release dateresearch.nvidia.com β†—2026-05-28verified
Architectureresearch.nvidia.com β†—2026-05-28verified
Total parametersresearch.nvidia.com β†—2026-05-28verified
Active parametersresearch.nvidia.com β†—2026-05-28verified
Training data cutoffbuild.nvidia.com β†—2026-05-28verified
Licensehuggingface.co β†—2026-05-28verified
MMLU-Proresearch.nvidia.com β†—2026-05-28verified
AGIEval-Enresearch.nvidia.com β†—2026-05-28verified
HumanEvalresearch.nvidia.com β†—2026-05-28verified
MBPP-Sanitizedresearch.nvidia.com β†—2026-05-28verified
GSM8Kresearch.nvidia.com β†—2026-05-28verified
MATHresearch.nvidia.com β†—2026-05-28verified
AIME 2025research.nvidia.com β†—2026-05-28verified
GPQA Diamondawesomeagents.ai β†—2026-05-28to verify
RULER (64K)llm-stats.com β†—2026-05-28to verify
NVIDIA Nemotron 3 family announced β€” Nano, Super, Ultraresearch.nvidia.com β†—2026-05-28verified
NVIDIA Technical Blog: Inside Nemotron 3 β€” Techniques, Tools, and Datadeveloper.nvidia.com β†—2026-05-28verified
Nemotron-3-Nano-Omni released β€” multimodal variant with audio+video understandinghuggingface.co/nvidia β†—2026-05-28verified
Nemotron-Cascade 2 β€” 30B open MoE runs one GPU, beats 120B modelsawesomeagents.ai β†—2026-05-28verified
What makes Nemotron 3 Nano different from other small MoE models?developer.nvidia.com β†—2026-05-28to verify
How does Nemotron 3 Nano compare to Qwen3-30B-A3B and GPT-OSS-20B?research.nvidia.com β†—2026-05-28to verify
Is Nemotron Nano good for agentic / tool-use applications?developer.nvidia.com β†—2026-05-28to verify
Is the Nemotron Nano model and its training data open source?research.nvidia.com β†—2026-05-28to verify
What pricing does TheRouter offer for Nemotron Nano?www.therouter.ai β†—2026-05-28to verify
Help & contact