GPT-4.1 vs GPT-5 Nano: pricing & cost comparison
On input tokens, GPT-5 Nano is the cheaper of the two — 98% less per million ($2 vs $0.05). On output, GPT-5 Nano is 95% cheaper ($8 vs $0.4) — and since output is usually the dominant cost driver, that gap matters more than it looks.
Side by side
| GPT-4.1 | GPT-5 Nano | |
|---|---|---|
| Input / 1M tokens | $2 | $0.05 |
| Output / 1M tokens | $8 | $0.4 |
| Context window | 1,047,576 | 400,000 |
| Token-count accuracy | exact | exact |
| Cost — 10,000 input + 2,000 output tokens | $0.036 | $0.0013 |
What a real request costs
Take a representative turn — 10,000 input + 2,000 output tokens. GPT-4.1 comes to $0.036, GPT-5 Nano to $0.0013. Across 100,000 requests that's a $3470 swing in favour of GPT-5 Nano. To run the numbers on your actual prompt, paste it into the calculator and toggle Compare across all models.
Which should you pick?
Both are OpenAI models, so you can move between them without changing SDKs or re-tokenising — route the routine 80% of traffic to the cheaper one and reserve GPT-4.1 for the genuinely hard requests. See the full breakdown on the dedicated pages for GPT-4.1 and GPT-5 Nano.
FAQ
- Is GPT-4.1 or GPT-5 Nano cheaper?
- For a typical request (10,000 input + 2,000 output tokens), GPT-5 Nano is cheaper — about 96% less, or roughly $3470 saved per 100,000 requests. GPT-4.1 runs $2/$8 per 1M input/output tokens; GPT-5 Nano runs $0.05/$0.4.
- Which has the larger context window?
- GPT-4.1, at 1,047,576 tokens versus 400,000.
- How accurate are these token counts?
- GPT-4.1: Exact tokenization via the canonical OpenAI vocab (o200k_base). GPT-5 Nano: Exact tokenization via the canonical OpenAI vocab (o200k_base). The dollar math itself is exact once the token count is known.