Amazon model product
Amazon Nova Pro is a highly-capable multimodal model with state-of-the-art performance across text, image, and video understanding.
Updated Aug 17, 2026. Default version: Nova Pro
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Benchmark scores for Nova Pro.
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCRAG | Score50.3% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkFinQA | Score77.2% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkGroundUI-1K | Score81.4% | Rank01 | Participants2 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkMM-Mind2Web | Score63.7% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkTranslation en→Set1 COMET22 | Score89.1% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkTranslation en→Set1 spBleu | Score43.4% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkTranslation Set1→en COMET22 | Score89.0% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkTranslation Set1→en spBleu | Score44.4% | Rank01 | Participants3 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkVisualWebBench | Score79.7% | Rank01 | Participants2 | Percentile100.0% | EvidenceC | Evaluated |
| BenchmarkVATEX | Score77.8% | Rank02 | Participants2 | Percentile0.0% | EvidenceC | Evaluated |
| BenchmarkBBH | Score86.9% | Rank03 | Participants12 | Percentile81.8% | EvidenceC | Evaluated |
| BenchmarkSQuALITY | Score19.8% | Rank03 | Participants5 | Percentile50.0% | EvidenceC | Evaluated |
| BenchmarkChartQA | Score89.2% | Rank04 | Participants26 | Percentile88.0% | EvidenceC | Evaluated |
| BenchmarkEgoSchema | Score72.1% | Rank04 | Participants9 | Percentile62.5% | EvidenceC | Evaluated |
| BenchmarkARC-C | Score94.8% | Rank05 | Participants34 | Percentile87.9% | EvidenceC | Evaluated |
| BenchmarkDROP | Score85.4% | Rank06 | Participants30 | Percentile82.8% | EvidenceC | Evaluated |
| BenchmarkTextVQA | Score81.5% | Rank07 | Participants16 | Percentile60.0% | EvidenceC | Evaluated |
| BenchmarkBFCL | Score68.4% | Rank08 | Participants11 | Percentile30.0% | EvidenceC | Evaluated |
| BenchmarkDocVQA | Score93.5% | Rank10 | Participants28 | Percentile66.7% | EvidenceC | Evaluated |
| BenchmarkIFEval | Score92.1% | Rank10 | Participants67 | Percentile86.4% | EvidenceC | Evaluated |
| BenchmarkGSM8k | Score94.8% | Rank13 | Participants48 | Percentile74.5% | EvidenceC | Evaluated |
| BenchmarkHumanEval | Score89.0% | Rank18 | Participants66 | Percentile73.8% | EvidenceC | Evaluated |
| BenchmarkLVBench | Score41.6% | Rank24 | Participants25 | Percentile4.2% | EvidenceC | Evaluated |
| BenchmarkMATH | Score76.6% | Rank25 | Participants71 | Percentile65.7% | EvidenceC | Evaluated |
| BenchmarkMMLU | Score85.9% | Rank36 | Participants101 | Percentile65.0% | EvidenceC | Evaluated |
| BenchmarkMMMU | Score61.7% | Rank43 | Participants63 | Percentile32.3% | EvidenceC | Evaluated |
| BenchmarkGPQA | Score46.9% | Rank191 | Participants239 | Percentile20.2% | EvidenceC | Evaluated |
Preference and agent-evaluation results for the default version.
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categoryoverall | Rank42 | Rating / score1447.3 | Votes3,409 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank53 | Rating / score1450.0 | Votes3,409 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank96 | Rating / score1420.1 | Votes3,708 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank99 | Rating / score1425.7 | Votes3,409 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank103 | Rating / score1424.1 | Votes3,525 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank111 | Rating / score1416.1 | Votes25,448 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank118 | Rating / score1415.2 | Votes3,422 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank119 | Rating / score1415.3 | Votes24,962 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank128 | Rating / score1404.6 | Votes3,422 | ObservationsN/A | Result date |
| Arenavision style control | Categoryoverall | Rank130 | Rating / score1019.5 | Votes2,335 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank132 | Rating / score1400.1 | Votes11,112 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank133 | Rating / score1398.3 | Votes3,422 | ObservationsN/A | Result date |
| Arenavision | Categoryoverall | Rank133 | Rating / score980.6 | Votes2,335 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank138 | Rating / score1394.7 | Votes11,536 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank148 | Rating / score1393.9 | Votes3,708 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank168 | Rating / score1364.5 | Votes2,836 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank178 | Rating / score1365.7 | Votes25,448 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank194 | Rating / score1349.6 | Votes11,536 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank199 | Rating / score1347.4 | Votes2,836 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank264 | Rating / score1258.8 | Votes24,745 | ObservationsN/A | Result date |
| Arenatext style control | Categoryoverall | Rank265 | Rating / score1290.1 | Votes24,745 | ObservationsN/A | Result date |
Official vendor API pricing appears first, followed by individual provider offers.
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderAmazon Bedrock | Provider model IDamazon.nova-pro-v1:0 | Regionglobal | Input / 1M$0.80 | Output / 1M$3.2 | Context300K | Updated |
| ProviderVercel AI Gateway | Provider model IDamazon/nova-pro | Regionglobal | Input / 1M$0.80 | Output / 1M$3.2 | Context300K | Updated |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Provider-specific output speed and catalog latency for Nova Pro. Runtime does not affect the capability score.
Provider | Output Speed | Catalog Latency | Max Input | Max Output | Updated |
|---|
| ProviderBedrock | Output Speed100 tok/s | Catalog Latency0.5 s | Max Input300K | Max Output300K | Updated |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Technical details for the model's default version.
Available versions of this model. The score column identifies the version used in the overall ranking.
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionNova Pro | Released | LLMBoard20.3 | ParametersN/A | Context300K | Max output300K | Open weightsNo | LicenseProprietary |
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
Key information about Nova Pro and its available data.
Amazon Nova Pro is a highly-capable multimodal model with state-of-the-art performance across text, image, and video understanding. It excels at core capabilities like language understanding, mathematical reasoning, and multimodal tasks while offering industry-leading speed and cost efficiency.
Data as of 2026-08-17.
Common questions about Nova Pro.
Nova Pro's default version was released on Nov 20, 2024.
Nova Pro's official API price is $0.80 per million input tokens and $3.2 per million output tokens via Amazon Bedrock. The lowest tracked third-party offer starts at $0.80 input and $3.2 output via Vercel AI Gateway.
Nova Pro was created by Amazon.
The default version has a 300K token context window.
No. The default version is not marked as having publicly available weights.
2 provider offerings are linked to the default version.
Nearby ranked alternatives include Pixtral Large, Gemini 2.0 Flash Lite, Grok 2.