Amazon API pricing
5 models tracked · headquartered in US · official pricing
Free tier: One-time credit What Amazon gives you free, in its own words
| Model | Input | Output | Context | |
|---|---|---|---|---|
| Nova Micro | $0.035 | $0.14 | 128K | reasoning vision |
| Nova Lite | $0.06 | $0.24 | 300K | reasoning vision |
| Nova 2 Lite | $0.3 | $2.5 | 1M | reasoning vision |
| Nova Pro | $0.8 | $3.2 | 300K | reasoning vision |
| Nova 2 Pro | $1.25 | $10 | 1M | reasoning vision |
Compare Amazon’s models on your own workload
How Amazon prices its models
Output prices across Amazon's 5 current models run from $0.14 per 1M tokens (Nova Micro) to $10 (Nova 2 Pro), a 71x spread. Picking the smallest model that does the job is usually the biggest saving available.
Discounts: caching and batch
5 of 5 current models list a cached-input rate, from 75% off the fresh input rate on Nova Micro. 0 list a batch discount for work that can wait.
Recorded price changes
- 2026-10-03: Nova 2 Pro input $2.1875 to $1.25
- 2026-10-03: Nova 2 Pro output $17.5 to $10
Where these prices come from
Every rate above is read from Amazon's own pricing page (linked here) and re-checked every morning. A disputed figure is marked rather than hidden; see how PromptSpend checks prices.
Common questions
How much do Amazon API models cost?
PromptSpend tracks 5 Amazon models. The cheapest on a blended rate is Nova Micro, at $0.035 per 1M input tokens and $0.14 per 1M output tokens; the most expensive is Nova 2 Pro, at $1.25 in and $10 out.
Does Amazon have a free API tier?
PromptSpend's verdict is "One-time credit". Its free-tier page gives Amazon's own wording for each claim, with a link to the page it came from and the date it was read.
Where do the Amazon prices come from?
From Amazon's own pricing page, https://aws.amazon.com/bedrock/pricing/, with each model's check date shown on its page. Prices change without notice, and regional premiums, priority tiers and negotiated discounts are not included.