Compare AI models
Claude Haiku 4.5vsgpt-5.4
Which should you pick?
- gpt-5.4 · overall 67.5 vs 54.6
- Claude Haiku 4.5 cheaper output · $5.00 vs $15.00 /1M
Specifications
| Claude Haiku 4.5 | gpt-5.4 | |
|---|---|---|
| Tier | A | A |
| Context window | 200K tokens | — |
| Input price (per 1M tokens) | $1.00 | $2.50 |
| Output price (per 1M tokens) | $5.00 | $15.00 |
| Provenance | Runs in:USMade in:United States | Runs in:USMade in:United States |
Quality by category
Win rate per category: how often each model beats a field-average model on a prompt from that category. 50% is average. It is not a percentage of correct answers.
| Category | Claude Haiku 4.5 | gpt-5.4 |
|---|---|---|
| coding | 60.4% | 65.1% |
| creative | 54.7% | 90.4% |
| factual | 49.8% | 57.1% |
| multilingual | 54.8% | 68.4% |
| reasoning | 58.4% | 74.2% |
Anthropic
Full review of Claude Haiku 4.5 →
OpenAI
Full review of gpt-5.4 →