Nemotron 4 Base: pricing, context window and benchmarks
Nemotron 4 Base is a open-weight model from NVIDIA. It offers a 8,192-token context window, costs $3.29 per million input tokens and $14.95 per million output tokens, and averages 73.0 across the index's benchmark suite.
Pricing
Nemotron 4 Base costs $3.29 per million input tokens and $14.95 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.21 per million tokens. Cached input is listed at $0.35 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.29 | per 1M tokens |
| Output | $14.95 | per 1M tokens |
| Cached input | $0.35 | per 1M tokens |
| Blended 3:1 | $6.21 | estimate per 1M tokens |
| 1M in + 1M out | $18.24 | 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 | 76.5 | 292 | 51th | |
| GPQA Diamond | 68.9 | 299 | 50th | |
| SWE-bench Verified | 60.7 | 312 | 48th | |
| HumanEval | 76.8 | 311 | 48th | |
| MATH-500 | 76.8 | 286 | 52th | |
| AIME 2025 | 66.7 | 320 | 47th | |
| LiveCodeBench | 64.2 | 313 | 48th | |
| IFEval | 81.4 | 294 | 51th | |
| TruthfulQA | 75.4 | 323 | 46th | |
| HellaSwag | 85.1 | 241 | 60th | |
| MMMU | 70.9 | 276 | 54th | |
| Arena Elo | 1364.5 | 333 | 45th |
What Nemotron 4 Base is good for
Nemotron 4 Base 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 $3.29 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.29 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
- Context window of 8,192 tokens is small for long-document work.
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Codestral Base | Mistral AI | 8192 | $3.31 | 73.1 |
| Mixtral Groq | Groq | 16384 | $3.37 | 73.1 |
| Claude 3.7 Sonnet Mini | Anthropic | 32768 | $9.74 | 73.0 |
| Llama 3.1 Turbo Base | Together AI | 16384 | $3.30 | 73.0 |
| Nemotron 4 Base | NVIDIA | 8192 | $3.29 | 73.0 |
| Mixtral Instruct | Mistral AI | 65536 | $3.35 | 72.9 |
FAQ
How much does Nemotron 4 Base cost?
Nemotron 4 Base is listed at $3.29 per million input tokens and $14.95 per million output tokens. Cached input is $0.35 per million tokens. A blended 3:1 input-to-output mix is about $6.21 per million tokens.
What is the context window of Nemotron 4 Base?
Nemotron 4 Base has a context window of 8,192 tokens and can emit up to 8,192 output tokens per request in this index.
Is Nemotron 4 Base open weights?
Yes. Nemotron 4 Base is listed as an open-weight model under a open weights licence.
How does Nemotron 4 Base score on benchmarks?
Nemotron 4 Base averages 73.0 across the scored benchmarks in this index, ranking #303 of 600 models. Its strongest relative result is Arena Elo at 1364.5.
What is Nemotron 4 Base best for?
Nemotron 4 Base 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 $3.29 per million input tokens (hosted list price) and a 8,192-token context window, it is a practical option for teams that want more control than a closed API allows.