Nova 2 Pro (preview) — API Pricing & Benchmarks

Nova 2 Pro (preview) is Amazon's flagship reasoning model, released in 2025-12. The API costs $1.25 per 1M input tokens and $10.0 per 1M output tokens ($3.44 blended at a 3:1 ratio), with cached input at $0.31. It supports a 1M-token context window. On the independently measured Artificial Analysis Intelligence Index it scores 14.2, ranking #57 of 72 models tracked on this site, with a median output speed of 116 tokens/sec.

Explore provider routes and additional benchmark sources for Nova 2 Pro (preview)

ProviderAmazon
Input price / 1M tokens$1.25
Output price / 1M tokens$10.0
Cached input / 1M tokens$0.31
Blended price (3:1)$3.44
Context window1M
Max output tokens66K
Knowledge cutoff—
Licenseproprietary
Input modalitiesT+I+A+V
Output speed (tok/s)116
Released2025-12
Artificial Analysis Intelligence Index14.2
GPQA Diamond78.5%
Humanity's Last Exam9.4%

Where Nova 2 Pro (preview) fits

On a blended 3:1 basis Nova 2 Pro (preview) costs $3.44 per 1M tokens, cheaper than 31% of the 84 priced models in this index. As Amazon's flagship it is aimed at the hardest workloads — complex reasoning, agentic coding and multi-step tool use — where output quality dominates cost. Cached input is priced at $0.31 (75% below standard input), which rewards prompt structures with long stable prefixes — see our caching guide.

Price history

PeriodInput $/1MOutput $/1MCached $/1M
2026-07-18 → current$1.25$10.0$0.31

No list-price changes recorded since tracking began. Full dataset: price-history.json (CC BY 4.0).

Alternatives at a similar price

About Amazon

Amazon's Nova models are built for AWS Bedrock, prioritizing price and throughput over peak benchmark scores. Nova Micro is among the cheapest models tracked in this index.

More Amazon models

← Full comparison table (60+ models, live filters)