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.
GPT-5 nano API pricing (September 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
- OpenAI
- Context window
- 400K tokens
- Max output
- 128K tokens
- Tier
- small
- Status
- GA
- Blended $/1M (3:1)
- $0.138
- Cache break-even hit rate
- 0%, caching always saves
- Prices verified
- 11 Sep 2026
Source: OpenAI pricing. Finitizer is not affiliated with OpenAI.
Similar models
Other gpt models
- GPT-5+$3.30
- GPT-5 mini+$0.550
- GPT-4.1+$3.36
- GPT-4.1 mini+$0.563
- GPT-4.1 nano+$0.038
- GPT-4o+$4.24
- GPT-4o mini+$0.125
- o3+$3.36
- o4-mini+$1.79
Closest prices elsewhere
- Amazon Nova Lite$-0.033
- Gemini 2.5 Flash-Lite+$0.038
- Gemini 2.0 Flash+$0.038
- DeepSeek V4 Flash+$0.038
Frequently asked questions
How much does GPT-5 nano cost per 1M tokens?
GPT-5 nano costs $0.050 per million input tokens and $0.400 per million output tokens at OpenAI list prices. Output is 8.0x the input rate, so the input:output ratio of your workload drives the real cost more than the headline price.
Does GPT-5 nano support prompt caching?
Yes. Cached input is billed at $0.00500 per million tokens, 10x cheaper than uncached input. There is no cache-write premium, so caching is never a net loss.
What is the batch discount for GPT-5 nano?
Requests submitted through the asynchronous batch endpoint are billed at 50% off both input and output rates, bringing input to $0.025 and output to $0.200 per million tokens.
What is GPT-5 nano's context window?
400K tokens of context with up to 128K 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 GPT-5 nano in production?
Finitizer TokenOps attributes real token spend to teams, features, and prompts, so you find out which workload moved the bill before the invoice does.