Is Claude 4.5 Haiku or GPT-5 Nano cheaper?

For a typical request (10,000 input + 2,000 output tokens), GPT-5 Nano is cheaper — about 94% less, or roughly $1870 saved per 100,000 requests. Claude 4.5 Haiku runs $1/$5 per 1M input/output tokens; GPT-5 Nano runs $0.05/$0.4.

Which has the larger context window?

GPT-5 Nano, at 400,000 tokens versus 200,000.

How accurate are these token counts?

Claude 4.5 Haiku: Approximated with cl100k_base — drift typically <2% on English and code. GPT-5 Nano: Exact tokenization via the canonical OpenAI vocab (o200k_base). The dollar math itself is exact once the token count is known.

Claude 4.5 Haiku vs GPT-5 Nano: pricing & cost comparison

On input tokens, GPT-5 Nano is the cheaper of the two — 95% less per million ($1 vs $0.05). On output, GPT-5 Nano is 92% cheaper ($5 vs $0.4) — and since output is usually the dominant cost driver, that gap matters more than it looks.

Side by side

	Claude 4.5 Haiku	GPT-5 Nano
Input / 1M tokens	$1	$0.05
Output / 1M tokens	$5	$0.4
Context window	200,000	400,000
Token-count accuracy	±2%	exact
Cost — 10,000 input + 2,000 output tokens	$0.02	$0.0013

What a real request costs

Take a representative turn — 10,000 input + 2,000 output tokens. Claude 4.5 Haiku comes to $0.02, GPT-5 Nano to $0.0013. Across 100,000 requests that's a $1870 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?

These are different vendors, so a switch means a different API and a slightly different tokenizer — budget a small calibration buffer. GPT-5 Nano give exact counts; the others land within a few percent. See the full breakdown on the dedicated pages for Claude 4.5 Haiku and GPT-5 Nano.

FAQ

Is Claude 4.5 Haiku or GPT-5 Nano cheaper?: For a typical request (10,000 input + 2,000 output tokens), GPT-5 Nano is cheaper — about 94% less, or roughly $1870 saved per 100,000 requests. Claude 4.5 Haiku runs $1/$5 per 1M input/output tokens; GPT-5 Nano runs $0.05/$0.4.
Which has the larger context window?: GPT-5 Nano, at 400,000 tokens versus 200,000.
How accurate are these token counts?: Claude 4.5 Haiku: Approximated with cl100k_base — drift typically <2% on English and code. GPT-5 Nano: Exact tokenization via the canonical OpenAI vocab (o200k_base). The dollar math itself is exact once the token count is known.