Claw Low vs GPT-4.1
Claw Low is cheaper on output tokens, while Claw Low offers a larger context window. Choose Claw Low or GPT-4.1 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Claw Low | GPT-4.1 |
|---|---|---|
| Provider | NanoGPT | OpenAI |
| Input / 1M tokens | $0.60 | $2.00 |
| Output / 1M tokens | $1.10 | $8.00 |
| Context window | 1M | 1M |
| Parameters | — | — |
| Open weights | No | No |
| Released | May 2026 | Apr 2025 |
| Aider Polyglot score | — | 52.4 |
| SWE-bench Multimodal score | — | 31.1 |
| FrontierMath score | — | 5.5 |
| Epoch Capabilities Index score | — | 137 |
| SWE-bench Verified score | — | 39.6 |
| GPQA Diamond score | — | 66.9 |
| LMArena (Chatbot Arena) Elo score | — | 1383 |
| LMArena Vision Elo score | — | 1211 |
FAQ
Is Claw Low or GPT-4.1 cheaper?
Claw Low is cheaper on output tokens ($1.10 vs $8.00 per 1M).
Which has the larger context window, Claw Low or GPT-4.1?
Claw Low 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.