Nova Pro Turbo: pricing, context window and benchmarks
Nova Pro Turbo is a multimodal model from Amazon. It offers a 128,000-token context window, costs $13.18 per million input tokens and $51.70 per million output tokens, and averages 78.4 across the index's benchmark suite.
Pricing
Nova Pro Turbo costs $13.18 per million input tokens and $51.70 per million output tokens. A blended 3:1 input-to-output mix works out to about $22.81 per million tokens. Cached input is listed at $1.33 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $13.18 | per 1M tokens |
| Output | $51.70 | per 1M tokens |
| Cached input | $1.33 | per 1M tokens |
| Blended 3:1 | $22.81 | estimate per 1M tokens |
| 1M in + 1M out | $64.88 | 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 | 81.8 | 128 | 79th | |
| GPQA Diamond | 72.4 | 212 | 65th | |
| SWE-bench Verified | 66.2 | 190 | 68th | |
| HumanEval | 82.5 | 176 | 71th | |
| MATH-500 | 81.1 | 190 | 68th | |
| AIME 2025 | 75.5 | 148 | 75th | |
| LiveCodeBench | 70.9 | 173 | 71th | |
| IFEval | 88.1 | 100 | 83th | |
| TruthfulQA | 83.5 | 94 | 84th | |
| HellaSwag | 86.1 | 215 | 64th | |
| MMMU | 74.4 | 175 | 71th | |
| Arena Elo | 1433.2 | 113 | 81th |
What Nova Pro Turbo is good for
Nova Pro Turbo 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 $13.18 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 128,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $13.18 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GLM-4.6 Base | Zhipu AI | 128000 | $4.53 | 78.5 |
| Reka Core Reasoning | Reka AI | 32768 | $13.11 | 78.5 |
| Magistral Instruct | Mistral AI | 65536 | $4.48 | 78.4 |
| Nova Pro Turbo | Amazon | 128000 | $13.18 | 78.4 |
| Qwen3-Coder Chat | Alibaba Cloud | 128000 | $4.59 | 78.4 |
| Reka Core Turbo | Reka AI | 32768 | $12.58 | 78.4 |
FAQ
How much does Nova Pro Turbo cost?
Nova Pro Turbo is listed at $13.18 per million input tokens and $51.70 per million output tokens. Cached input is $1.33 per million tokens. A blended 3:1 input-to-output mix is about $22.81 per million tokens.
What is the context window of Nova Pro Turbo?
Nova Pro Turbo has a context window of 128,000 tokens and can emit up to 4,096 output tokens per request in this index.
Is Nova Pro Turbo open weights?
No. Nova Pro Turbo is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Nova Pro Turbo score on benchmarks?
Nova Pro Turbo averages 78.4 across the scored benchmarks in this index, ranking #158 of 600 models. Its strongest relative result is Arena Elo at 1433.2.
What is Nova Pro Turbo best for?
Nova Pro Turbo 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 $13.18 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.