Nemotron 4 8B: pricing, context window and benchmarks
Nemotron 4 8B is a small open-weight model from NVIDIA. It offers a 4,096-token context window, costs $1.06 per million input tokens and $5.67 per million output tokens, and averages 57.9 across the index's benchmark suite.
Pricing
Nemotron 4 8B costs $1.06 per million input tokens and $5.67 per million output tokens. A blended 3:1 input-to-output mix works out to about $2.21 per million tokens. Cached input is listed at $0.10 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $1.06 | per 1M tokens |
| Output | $5.67 | per 1M tokens |
| Cached input | $0.10 | per 1M tokens |
| Blended 3:1 | $2.21 | estimate per 1M tokens |
| 1M in + 1M out | $6.73 | 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 | 63.8 | 475 | 21th | |
| GPQA Diamond | 50.8 | 514 | 14th | |
| SWE-bench Verified | 40.9 | 512 | 15th | |
| HumanEval | 61.2 | 484 | 19th | |
| MATH-500 | 60.0 | 465 | 23th | |
| AIME 2025 | 51.2 | 449 | 25th | |
| LiveCodeBench | 46.9 | 477 | 21th | |
| IFEval | 68.8 | 480 | 20th | |
| TruthfulQA | 65.5 | 454 | 24th | |
| HellaSwag | 71.7 | 506 | 16th | |
| MMMU | 55.6 | 492 | 18th | |
| Arena Elo | 1247.2 | 495 | 18th |
What Nemotron 4 8B is good for
Nemotron 4 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $1.06 per million input tokens it can handle a large share of production traffic.
Strengths
- Input price of $1.06 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
Trade-offs
- Context window of 4,096 tokens is small for long-document work.
- Average benchmark score is below the index median; test it on your own task.
- Less reliable on open-ended reasoning and complex multi-step instructions.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Firefunction Nano | Fireworks AI | 16384 | $3.07 | 58.1 |
| Nemotron 4 8B | NVIDIA | 4096 | $1.06 | 57.9 |
| Llama 3.3 7B | Meta AI | 8192 | $1.18 | 57.8 |
| Command R+ 8B | Cohere | 8192 | $1.21 | 57.7 |
| Hunyuan-Large 7B | Tencent | 32768 | $1.08 | 57.7 |
| MiniMax-Text 7B | MiniMax | 16384 | $1.10 | 57.7 |
FAQ
How much does Nemotron 4 8B cost?
Nemotron 4 8B is listed at $1.06 per million input tokens and $5.67 per million output tokens. Cached input is $0.098 per million tokens. A blended 3:1 input-to-output mix is about $2.21 per million tokens.
What is the context window of Nemotron 4 8B?
Nemotron 4 8B has a context window of 4,096 tokens and can emit up to 4,096 output tokens per request in this index.
Is Nemotron 4 8B open weights?
Yes. Nemotron 4 8B is listed as an open-weight model under a open weights licence.
How does Nemotron 4 8B score on benchmarks?
Nemotron 4 8B averages 57.9 across the scored benchmarks in this index, ranking #474 of 600 models. Its strongest relative result is Arena Elo at 1247.2.
What is Nemotron 4 8B best for?
Nemotron 4 8B is a small model. That makes it a good fit for classification, extraction, routing, summarization and other high-volume jobs where latency and cost dominate. It is not the right pick for open-ended frontier reasoning, but at $1.06 per million input tokens it can handle a large share of production traffic.