Comparison · prices as of 2026-07-04 · re-verified 2026-08-18

Claude Haiku 4.5 vs GPT-5 nano

On price alone, GPT-5 nano (OpenAI) runs about 93% cheaper than Claude Haiku 4.5 for a typical chat workload (2K input + 500 output tokens, 1,000 requests/day). Whether that's the right trade depends on the capability gap — the breakdown below shows where each one wins.

Price per million tokens

Spec Claude Haiku 4.5 GPT-5 nano
Input / MTok $1 $0.05
Output / MTok $5 $0.40
Context window 200K 400K
Max output 64K 128K
Class fast / lightweight fast / lightweight

Estimated monthly cost by workload

Same models, three realistic usage levels. The gap between them scales with volume, so the right pick can change as you grow.

Workload Claude Haiku 4.5 GPT-5 nano Cheaper
Light — chatbot 0.5K in / 0.2K out · 200 req/day $9.00 $0.63 GPT-5 nano
Typical — assistant 2K in / 0.5K out · 1K req/day $135 $9.00 GPT-5 nano
Heavy — RAG / agent 8K in / 2K out · 5K req/day $2,700 $180 GPT-5 nano

Which should you pick?

Frequently asked questions

Is Claude Haiku 4.5 or GPT-5 nano cheaper?

GPT-5 nano is cheaper — roughly 93% less per month on a typical assistant workload (2K input + 500 output tokens, 1,000 requests/day).

What do Claude Haiku 4.5 and GPT-5 nano cost per million tokens?

Claude Haiku 4.5 is $1 per million input tokens and $5 per million output. GPT-5 nano is $0.05 input and $0.40 output. Prices are first-party API list rates as of 2026-07-04.

Which has the larger context window, Claude Haiku 4.5 or GPT-5 nano?

GPT-5 nano does, at 400K tokens versus 200K for Claude Haiku 4.5.

When should I choose Claude Haiku 4.5 over GPT-5 nano?

Both are fast / lightweight-class models, so choose Claude Haiku 4.5 when it measurably wins on your own evals or you prefer its provider (Anthropic); otherwise GPT-5 nano does the same job for less.

Related comparisons

Sources: Anthropic pricing · OpenAI pricing. Run your own numbers in the cost calculator.