Claude Haiku 4.5 vs GPT-5.4
A direct API-pricing comparison between Claude Haiku 4.5 and GPT-5.4. Below: every cost dimension side by side, then the bottom line for a typical workload.
Bottom line
For a typical 5K-in / 1K-out request, Claude Haiku 4.5 is the cheaper choice at $0.010 per call — 2.8× less than GPT-5.4 at $0.028.
| Claude Haiku 4.5Anthropic | GPT-5.4OpenAI | |
|---|---|---|
| Input / 1M tokens | $1.00 | $2.50 |
| Output / 1M tokens | $5.00 | $15.00 |
| Cached input / 1M | $0.100 | — |
| Context window | 200K | 400K |
| Provider | Anthropic | OpenAI |
Green = cheaper on that line. Prices are estimates from public pricing pages; verify against Anthropic and OpenAI before relying on them.
Which is cheaper, Claude Haiku 4.5 or GPT-5.4?
On output tokens — usually the dominant cost — Claude Haiku 4.5 charges $5.00 per million versus $15.00 for GPT-5.4. The right pick depends on your input/output ratio and how much you can serve from cache. Use the calculator to model your exact traffic.