Claude 3 Haiku vs GPT-4.1
Side-by-side API pricing and Artificial Analysis performance for Claude 3 Haiku (Anthropic) and GPT-4.1 (OpenAI). Green cells mark the better value in each row.
Prices updated Jul 22, 2026, 1:38 AM UTC · refreshed hourly
| Claude 3 Haiku Anthropic | GPT-4.1 OpenAI | |
|---|---|---|
| Pricing | ||
| Input $/1M | $0.25 | $2.00 |
| Output $/1M | $1.25 | $8.00 |
| Blended $/1M (3:1) | $0.50 | $3.50 |
| RAG example (30K in / 2K out) | $0.01 | $0.076 |
| Quality & speed | ||
| Intelligence Index | 3.9 | 19.4 |
| Coding Index | — | — |
| Agentic Index | — | — |
| Output speed (tok/s) | 0 | 149 |
| Time to first token | 0s | 0.56s |
| Specs | ||
| Context window | 200K | 1.05M |
| Input modalities | text, image | image, text, file |
FAQ
Which is cheaper, Claude 3 Haiku or GPT-4.1?
Claude 3 Haiku has the lower blended API price at $0.50 per 1M tokens (3:1 input:output mix), versus $3.50 for the other model. Prices are live from OpenRouter and refresh about hourly.
Which is better for RAG workloads?
For a typical RAG request (30K input / 2K output tokens), Claude 3 Haiku costs about $0.01 per request versus $0.076. Also compare context windows: Claude 3 Haiku offers 200K and GPT-4.1 offers 1.05M.
Which scores higher on Artificial Analysis benchmarks?
GPT-4.1 leads on the Artificial Analysis Intelligence Index (19.4 vs 3.9). Check coding and agentic indices on this page for workload-specific tradeoffs.