OpenAI-compatible — swap the base URL and it works with your existing SDK.
| Type | Price |
|---|---|
| Input | |
| Output | |
| Benchmark | Score | What it measures | Source | Retrieved |
|---|---|---|---|---|
| Intelligence | 7.4 | Composite score across standardised reasoning, knowledge and problem-solving evaluations, measured independently by Artificial Analysis | Source | September 22, 2026 |
| Math | 69.7 | Composite score across standardised mathematics evaluations, measured independently by Artificial Analysis | Source | September 22, 2026 |
| AIME 2025 | 72.1% | Competition mathematics (AIME), 2025 | Source | September 22, 2026 |
| GPQA Diamond | 64.0% | Google-proof graduate science questions (hardest subset) | Source | September 22, 2026 |
| LiveCodeBench | 71.1% | Contamination-free competitive programming problems | Source | September 22, 2026 |
| Humanity's Last Exam | 6.5% | Expert-level questions across many domains | Source | September 22, 2026 |
| Model | Input | Output | Context | Best for |
|---|---|---|---|---|
Nemotron Nano 9B V2 This page | ||||
| Reasoning + agents | ||||
| Balanced coding + agents | ||||
| Long-context, multimodal & agentic workflows | ||||
| Reasoning + agents |
Yes, Nemotron Nano 9B V2 can stream responses as they are generated.
Yes, Nemotron Nano 9B V2 supports both function calling and structured outputs.
Nemotron Nano 9B V2 was built by NVIDIA.
No. Its listed capabilities are function calling, streaming, and structured outputs, with no image input support mentioned.
Requests are sent to the v1/chat/completions endpoint.
