Amazon Nova Pro is AWS's flagship model in the Nova family, tightly integrated with the AWS Bedrock platform and optimized for enterprise AWS workloads. Nova Pro supports multimodal inputs (text, images, video, documents), a 300K-token context window, tool use, and streaming. It is designed for the broad middle ground of enterprise tasks — strong enough for complex workloads, fast enough for interactive applications, and priced competitively for high-volume AWS deployments. It is available exclusively through AWS Bedrock.
Amazon Nova Pro is priced at $0.80 per million input tokens and $3.20 per million output tokens — significantly cheaper than comparable advanced-tier models. At 1,000 input + 500 output tokens per call, it costs approximately $0.0024 per 1,000 API calls. For teams already on AWS with high-volume workloads, the cost advantage is substantial. AWS Bedrock's pricing also includes volume discounts and committed use options that further reduce costs for large-scale deployments.
Amazon Nova Pro's 300K-token context window is larger than most competitors at its price point, enabling analysis of very large documents and codebases. Median latency is approximately 300–500ms TTFT. Throughput is approximately 100 tokens per second. Maximum output is 5,120 tokens — lower than many competing models but adequate for most enterprise tasks. Video understanding (up to 30 minutes of video) and document processing (PDFs, spreadsheets) are native capabilities, making it well-suited to enterprise document workflows. The model is only available through AWS Bedrock, which simplifies compliance and security for AWS-native organizations.
Amazon Nova Pro is best suited to enterprise AWS workloads requiring multimodal capability, large context windows, and competitive pricing: document analysis, video summarization, enterprise Q&A, content generation, and agentic workflows within the AWS ecosystem. Its tight Bedrock integration enables seamless use of AWS services (S3, Lambda, Kendra) alongside the model. For teams already invested in AWS infrastructure, Nova Pro's combination of competitive pricing, large context window, and multimodal capability makes it a strong default choice. For teams not on AWS, the Bedrock-only availability is a significant constraint.
Amazon Nova Pro sits in the middle of Amazon's Nova lineup, above Nova Lite (faster, cheaper, less capable) and below Nova Premier (most capable, most expensive). Its main competitors are advanced-tier proprietary models from OpenAI and Anthropic (more expensive, broader ecosystems), and efficient-tier models from Google (cheaper, less capable). For AWS-native teams, Nova Pro is the natural default for new applications — it offers the best cost-to-capability ratio within the Bedrock ecosystem. Its intelligence score of approximately 78/100 reflects solid advanced-tier performance, particularly for enterprise document and multimodal tasks.
Composite quality score based on reasoning, coding, and instruction-following benchmarks.
Input price vs 184 advanced models tracked. Avg: $1.38/1M.
| Model ID | amazon.nova-pro-v1:0 |
| Provider | Amazon Bedrock |
| Model Family | Amazon Nova |
| Tier | Advanced |
| Context Window | 300,000 tokens (300K) |
| Max Output Tokens | 5,120 tokens |
| Input Price | $0.8 / 1M tokens |
| Output Price | $3.2 / 1M tokens |
| Cached Input Price | Not available |
| Throughput | 130 tokens/sec |
| Latency (P50) | 500ms |
| Intelligence Score | 60/100 |
| LiveBench Score | — |
| Cost per Task (LiveBench) | — |
| Data Confidence | — |
| HLE Score | — |
| GPQA Diamond | — |
| SWE-bench Score | — |
| OSWorld Score | — |
| BrowseComp Score | — |
| Terminal-Bench Score | — |
| AutoBench Score | — |
| Parameters | — |
| Vision Support | Yes |
| Function Calling | Yes |
| Streaming | Yes |
| Reasoning Model | No |
| Open Weights | No |
| Data Source | Static (verified) |
| Last Updated | 2026-07-20 |
Similar Models by Intelligence Score
Self-hosting amazon.nova-pro-v1:0? These GPUs are well-matched for this model's compute and memory requirements.