Nova Lite Reasoning: pricing, context window and benchmarks
Nova Lite Reasoning is a multimodal model from Amazon. It offers a 32,768-token context window, costs $9.92 per million input tokens and $53.42 per million output tokens, and averages 74.3 across the index's benchmark suite.
Pricing
Nova Lite Reasoning costs $9.92 per million input tokens and $53.42 per million output tokens. A blended 3:1 input-to-output mix works out to about $20.80 per million tokens. Cached input is listed at $1.42 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $9.92 | per 1M tokens |
| Output | $53.42 | per 1M tokens |
| Cached input | $1.42 | per 1M tokens |
| Blended 3:1 | $20.80 | estimate per 1M tokens |
| 1M in + 1M out | $63.34 | 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 | 78.7 | 223 | 63th | |
| GPQA Diamond | 69.0 | 296 | 51th | |
| SWE-bench Verified | 62.4 | 270 | 55th | |
| HumanEval | 80.4 | 225 | 63th | |
| MATH-500 | 78.6 | 248 | 59th | |
| AIME 2025 | 66.7 | 321 | 47th | |
| LiveCodeBench | 65.7 | 281 | 53th | |
| IFEval | 83.5 | 233 | 61th | |
| TruthfulQA | 76.8 | 286 | 52th | |
| HellaSwag | 84.0 | 282 | 53th | |
| MMMU | 72.0 | 247 | 59th | |
| Arena Elo | 1371.0 | 313 | 48th |
What Nova Lite Reasoning is good for
Nova Lite Reasoning is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $9.92 per million input tokens with a 32,768-token window, check its benchmark profile against the specific task before committing.
Strengths
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $9.92 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Llama 3 Novita | Novita AI | 128000 | $3.70 | 74.4 |
| Seed-OSS Chat | ByteDance | 65536 | $3.49 | 74.4 |
| GPT-4.1 Mini | OpenAI | 128000 | $10.43 | 74.3 |
| Nova Lite Reasoning | Amazon | 32768 | $9.92 | 74.3 |
| Gemini 2.5 Pro Mini | Google DeepMind | 128000 | $10.07 | 74.2 |
| Llama 3 Novita Chat | Novita AI | 65536 | $3.50 | 74.2 |
FAQ
How much does Nova Lite Reasoning cost?
Nova Lite Reasoning is listed at $9.92 per million input tokens and $53.42 per million output tokens. Cached input is $1.42 per million tokens. A blended 3:1 input-to-output mix is about $20.80 per million tokens.
What is the context window of Nova Lite Reasoning?
Nova Lite Reasoning has a context window of 32,768 tokens and can emit up to 32,768 output tokens per request in this index.
Is Nova Lite Reasoning open weights?
No. Nova Lite Reasoning is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Nova Lite Reasoning score on benchmarks?
Nova Lite Reasoning averages 74.3 across the scored benchmarks in this index, ranking #269 of 600 models. Its strongest relative result is Arena Elo at 1371.0.
What is Nova Lite Reasoning best for?
Nova Lite Reasoning is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $9.92 per million input tokens with a 32,768-token window, check its benchmark profile against the specific task before committing.