Nemotron 4 Instruct: pricing, context window and benchmarks
Nemotron 4 Instruct is a multimodal model from NVIDIA. It offers a 65,536-token context window, costs $4.10 per million input tokens and $14.46 per million output tokens, and averages 76.4 across the index's benchmark suite.
Pricing
Nemotron 4 Instruct costs $4.10 per million input tokens and $14.46 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.69 per million tokens. Cached input is listed at $0.59 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.10 | per 1M tokens |
| Output | $14.46 | per 1M tokens |
| Cached input | $0.59 | per 1M tokens |
| Blended 3:1 | $6.69 | estimate per 1M tokens |
| 1M in + 1M out | $18.56 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 81.3 | 140 | 77th | |
| GPQA Diamond | 74.2 | 162 | 73th | |
| SWE-bench Verified | 64.4 | 224 | 63th | |
| HumanEval | 83.3 | 154 | 74th | |
| MATH-500 | 77.9 | 260 | 57th | |
| AIME 2025 | 69.8 | 261 | 57th | |
| LiveCodeBench | 66.4 | 266 | 56th | |
| IFEval | 83.0 | 245 | 59th | |
| TruthfulQA | 80.7 | 164 | 73th | |
| HellaSwag | 84.9 | 253 | 58th | |
| MMMU | 74.2 | 181 | 70th | |
| Arena Elo | 1413.2 | 177 | 71th |
What Nemotron 4 Instruct is good for
Nemotron 4 Instruct is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $4.10 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.
Strengths
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Claude Sonnet 4.5 Mini | Anthropic | 32768 | $11.43 | 76.5 |
| Qwen3-Max Mini | Alibaba Cloud | 128000 | $11.13 | 76.5 |
| Yi-Lightning Instruct | 01.AI | 128000 | $4.16 | 76.5 |
| MiniMax-M2 Base | MiniMax | 65536 | $4.20 | 76.4 |
| Nemotron 4 Instruct | NVIDIA | 65536 | $4.10 | 76.4 |
| Llama 3.1 | Meta AI | 200000 | $4.16 | 76.3 |
FAQ
How much does Nemotron 4 Instruct cost?
Nemotron 4 Instruct is listed at $4.10 per million input tokens and $14.46 per million output tokens. Cached input is $0.59 per million tokens. A blended 3:1 input-to-output mix is about $6.69 per million tokens.
What is the context window of Nemotron 4 Instruct?
Nemotron 4 Instruct has a context window of 65,536 tokens and can emit up to 65,536 output tokens per request in this index.
Is Nemotron 4 Instruct open weights?
Yes. Nemotron 4 Instruct is listed as an open-weight model under a open weights licence.
How does Nemotron 4 Instruct score on benchmarks?
Nemotron 4 Instruct averages 76.4 across the scored benchmarks in this index, ranking #210 of 600 models. Its strongest relative result is Arena Elo at 1413.2.
What is Nemotron 4 Instruct best for?
Nemotron 4 Instruct is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $4.10 per million input tokens, it sits in the mid-range of the index, and the 65,536-token context window is enough for large files and multi-file diffs.