Nemotron 4 Chat: pricing, context window and benchmarks
Nemotron 4 Chat is a multimodal model from NVIDIA. It offers a 65,536-token context window, costs $3.99 per million input tokens and $16.11 per million output tokens, and averages 75.0 across the index's benchmark suite.
Pricing
Nemotron 4 Chat costs $3.99 per million input tokens and $16.11 per million output tokens. A blended 3:1 input-to-output mix works out to about $7.02 per million tokens. Cached input is listed at $0.45 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.99 | per 1M tokens |
| Output | $16.11 | per 1M tokens |
| Cached input | $0.45 | per 1M tokens |
| Blended 3:1 | $7.02 | estimate per 1M tokens |
| 1M in + 1M out | $20.10 | 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 | 77.2 | 271 | 55th | |
| GPQA Diamond | 69.4 | 283 | 53th | |
| SWE-bench Verified | 62.3 | 271 | 55th | |
| HumanEval | 78.6 | 277 | 54th | |
| MATH-500 | 82.2 | 160 | 73th | |
| AIME 2025 | 73.0 | 203 | 66th | |
| LiveCodeBench | 65.9 | 276 | 54th | |
| IFEval | 81.8 | 281 | 53th | |
| TruthfulQA | 79.2 | 210 | 65th | |
| HellaSwag | 84.4 | 268 | 55th | |
| MMMU | 71.1 | 271 | 55th | |
| Arena Elo | 1398.2 | 212 | 65th |
What Nemotron 4 Chat is good for
Nemotron 4 Chat is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.99 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.99 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
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 |
|---|---|---|---|---|
| Codestral Instruct | Mistral AI | 65536 | $4.01 | 75.1 |
| Llama 3 Hyperbolic Chat | Hyperbolic | 200000 | $3.64 | 75.1 |
| Llama 3 Replicate Instruct | Replicate | 128000 | $3.71 | 75.1 |
| Mistral Small 3 Instruct | Mistral AI | 32768 | $3.97 | 75.1 |
| Llama 3 DeepInfra Chat | DeepInfra | 200000 | $3.74 | 75.0 |
| Nemotron 4 Chat | NVIDIA | 65536 | $3.99 | 75.0 |
FAQ
How much does Nemotron 4 Chat cost?
Nemotron 4 Chat is listed at $3.99 per million input tokens and $16.11 per million output tokens. Cached input is $0.45 per million tokens. A blended 3:1 input-to-output mix is about $7.02 per million tokens.
What is the context window of Nemotron 4 Chat?
Nemotron 4 Chat 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 Chat open weights?
Yes. Nemotron 4 Chat is listed as an open-weight model under a open weights licence.
How does Nemotron 4 Chat score on benchmarks?
Nemotron 4 Chat averages 75.0 across the scored benchmarks in this index, ranking #254 of 600 models. Its strongest relative result is Arena Elo at 1398.2.
What is Nemotron 4 Chat best for?
Nemotron 4 Chat is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.99 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.