GPT-4.1 vs GPT-5
GPT-4.1 is cheaper on output tokens, while GPT-4.1 offers a larger context window. Choose GPT-4.1 or GPT-5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | GPT-4.1 | GPT-5 |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Input / 1M tokens | $2.00 | $1.25 |
| Output / 1M tokens | $8.00 | $10.00 |
| Context window | 1M | 400K |
| Parameters | — | — |
| Open weights | No | No |
| Released | Apr 2025 | Aug 2025 |
| Aider Polyglot score | 52.4 | 88.0 |
| GPQA Diamond score | 66.9 | 86.2 |
| FrontierMath score | 5.5 | 32.4 |
| SWE-bench Verified score | 39.6 | 75.6 |
| SWE-bench Multimodal score | 31.1 | — |
| Epoch Capabilities Index score | 137 | 150 |
| LMArena (Chatbot Arena) Elo score | 1383 | 1406 |
| LMArena Vision Elo score | 1211 | 1208 |
| LMArena WebDev Elo score | — | 1417 |
FAQ
Is GPT-4.1 or GPT-5 cheaper?
GPT-4.1 is cheaper on output tokens ($8.00 vs $10.00 per 1M).
Which has the larger context window, GPT-4.1 or GPT-5?
GPT-4.1 has the larger context window (1M tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.