AmazonNova Pro
Nova Pro is an Amazon proprietary model, released 3 Dec 2024. Its best configuration ranks #671 of 785 on the BenchLeader Index at 39.8 ±3.6, in the lower half. The ± is the point: 242 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in instruction following (46) and lowest in multimodal (30). At $1.40 per million tokens blended it is pricier than most ranked models. Output speed of 129 tokens per second puts it faster than most, with a first answer in 1.1 s. Last measured 24 Nov 2025.stale: no new result in six months
- Blended price
- $1.40/M
- $0.800 in · $3.20 out
- Output speed
- 129 tok/s
- measured by Artificial Analysis
- First answer
- 1.11 s
- first token 0.70 s
- Context
- 300k
- Full answer
- 4.99 s
- median, reasoning included
- Cost per run
- –
- Released
- 3 Dec 2024
- Overall index40
- Reasoning37
- Agents & tools38
- Maths38
- Knowledge40
- Instruction following46
- Multimodal30
- Long context34
- Composite37
Versions
Amazon has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Nova Prothis page | 3 Dec 2024 | 39.8 | #671 |
| Nova Pro 1.0 | – | 43.8 | #560 |
How its index has moved
9 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source |
|---|---|---|
| Humanity's Last Exam | 4.4%#45 | Scale AI / CAIS |
| GPQA Diamond (AA)not in index | 49.9%#430 | Artificial Analysis |
| Humanity's Last Exam (AA)not in index | 3.2%#602 | Artificial Analysis |
| GPQA Diamond (HELM)not in index | 44.6%#40 | HELM Capabilities |
| CritPt | 0.0%#315 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source |
|---|---|---|
| Terminal-Bench Hard | 6.1%#266 | Artificial Analysis |
| τ²-Bench Telecom (AA)not in index | 14.0%#367 | Artificial Analysis |
| BFCL Overall | 25.0%#66 | Berkeley Function Calling Leaderboard |
| TheAgentCompanynot in index | 1.7%#13 | TheAgentCompany |
Maths
| Benchmark | not stated | Source |
|---|---|---|
| Omni-MATH (HELM) | 24.2%#52 | HELM Capabilities |
Knowledge
| Benchmark | not stated | Source |
|---|---|---|
| AA-Omniscience | -47.7#369 | Artificial Analysis |
| MMLU-Pro (HELM)not in index | 67.3%#40 | HELM Capabilities |
| MMLUnot in index | 82.0%#15 | Epoch AI Benchmarking Hub |
| AA-Omniscience: accuracynot in index | 16.9%#379 | Artificial Analysis |
| AA-Omniscience: non-hallucinationnot in index | 22.3%#266 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source |
|---|---|---|
| IFBench | 38.1%#282 | Artificial Analysis |
| IFEval (HELM) | 81.5%#41 | HELM Capabilities |
| WildBench (HELM) | 77.7%#48 | HELM Capabilities |
Human preference
| Benchmark | not stated | Source |
|---|---|---|
| Lech Mazur Creative Writingnot in index | 60.5%#42 | Lech Mazur |
Multimodal
| Benchmark | not stated | Source |
|---|---|---|
| MMMU-Pro | 44.3%#257 | Artificial Analysis |
| VISTA | 26.3%#52 | Scale AI SEAL |
| MMMU (validation) | 62.0%#31 | MMMU |
Long context
| Benchmark | not stated | Source |
|---|---|---|
| AA-LCR | 21.0%#428 | Artificial Analysis |
Composite
| Benchmark | not stated | Source |
|---|---|---|
| AA Intelligence Index v4.3.2 | 7.0#498 | Artificial Analysis |
| HELM Capabilities meannot in index | 59.1%#46 | HELM Capabilities |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Amazon Bedrock | 74 tok/s | 0.66 s | $0.800 | $3.20 | 300k | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.200 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0013 | $0.0011 | 3.4 s |
| Summarize a 30-page report | 12,000 / 600 | $0.011 | $0.0061 | 5.8 s |
| Code edit | 6,000 / 1,500 | $0.0096 | $0.0069 | 12.7 s |
| Agentic coding session | 60,000 / 4,000 | $0.061 | $0.034 | 32.1 s |
| Structured extraction | 2,000 / 200 | $0.0022 | $0.0013 | 2.7 s |
See also
Questions people ask
- What are Nova Pro's benchmark scores?
- As of 25 Oct 2026, Nova Pro scores 39.8 ±3.6 on the BenchLeader Index, strongest in instruction following (46) and weakest in multimodal (30). That rests on 14 independent benchmark results, and every score on this page links to its publisher.
- How does Nova Pro rank?
- Nova Pro is #671 of 785 ranked configurations on 25 Oct 2026. 242 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Nova Pro cost per 1M tokens?
- As of 25 Oct 2026, Nova Pro costs $0.800 per 1M input tokens and $3.20 per 1M output tokens, $1.40 blended (three input tokens per output token).
- When was Nova Pro released?
- Amazon released Nova Pro on 3 Dec 2024.
Data as of 25 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Nova Pro: benchmarks, pricing, speed and rank”, data as of 25 Oct 2026, https://www.benchleader.com/models/nova-pro.