A rate per million tokens is not a budget. The calculator takes your request volume, prompt size and cache hit rate and gives you a monthly figure.
Amazon Nova Premier API pricing (October 2026)
What it costs at real volumes
A chat-shaped workload: 800 input tokens and 300 output tokens per request, no caching, no batch.
Specifications
- Provider
- Amazon Bedrock
- Context window
- 1.0M tokens
- Max output
- 32K tokens
- Tier
- Flagship
- Status
- GA
- Blended $/1M (3:1)
- $5.00
- Cache break-even, caching each turn
- 0%, caching always saves
- Prices verified
- 30 Oct 2026
Source: Amazon Bedrock pricing. Finitizer is not affiliated with Amazon Bedrock.
Similar models
Other Nova models
- Amazon Nova 2 Lite-$4.15
- Amazon Nova Pro-$3.60
- Amazon Nova Lite-$4.90
- Amazon Nova Micro-$4.94
Closest prices elsewhere
- GPT-5.6 Terra-$0.500
- Gemini 3.1 Pro Preview-$0.500
- Claude Sonnet 5.5-$1.00
- GPT-6.1 Sol-$1.00
Frequently asked questions
How much does Amazon Nova Premier cost per 1M tokens?
Amazon Nova Premier costs $2.50 per million input tokens and $12.50 per million output tokens at Amazon Bedrock list prices. Output is 5.0x the input rate, so the input:output ratio of a workload drives the real cost more than the headline price.
Does Amazon Nova Premier support prompt caching?
Yes. Cached input is billed at $0.625 per million tokens, 4x cheaper than uncached input. There is no cache-write premium, so caching is never a net loss.
What is the batch discount for Amazon Nova Premier?
Requests submitted through the asynchronous batch endpoint are billed at 50% off both input and output rates, bringing input to $1.25 and output to $6.25 per million tokens.
What is Amazon Nova Premier's context window?
1.0M tokens of context with up to 32K output tokens per request. A larger context window raises the ceiling on what you can send, not the price per token, but filling it on every request is one of the most common causes of an unexpected bill.
Running Amazon Nova Premier in production?
Finitizer Tokenomics reads what this model actually costs you from your own bills and exports, shows which workload, team or API key is spending it, flags a spike or a new spend source the day your data shows it, and forecasts where the month will land.
See Finitizer Tokenomics