rs/curlie
The power of curl, the ease of use of httpie.
Results by Model
| # | Model | Score Percentage of hidden behavioral tests passed. | Cost Total API cost in USD for this task instance. | Calls Number of LLM API calls for this task instance. | |
|---|---|---|---|---|---|
| 1 |  |
Claude Opus 4.7 (xhigh) Anthropic | 92.9% | $18.14 | 157 |
| 2 |  |
GPT 5.5 (xhigh) OpenAI | 92.8% | $7.88 | 81 |
| 3 | |
Claude Opus 4.6 Anthropic | 89.3% | $12.49 | 243 |
| 4 | |
Claude Sonnet 4.6 Anthropic | 88.3% | $25.25 | 518 |
| 5 | |
GPT 5.5 (high) OpenAI | 87.7% | $3.25 | 38 |
| 6 | |
Claude Opus 4.7 Anthropic | 84.8% | $15.70 | 247 |
| 7 | |
GPT 5.5 OpenAI | 83.1% | $1.21 | 23 |
| 8 |  |
Gemini 3.1 Pro Google | 79.8% | $1.62 | 127 |
| 9 | |
Gemini 3 Flash Google | 71.1% | $0.22 | 85 |
| 10 | |
GPT 5.4 OpenAI | 54.0% | $0.27 | 10 |
| 11 | |
Claude Haiku 4.5 Anthropic | 43.0% | $0.59 | 118 |
| 12 | |
GPT 5.4 mini OpenAI | 11.3% | $0.02 | 9 |
| 13 | |
GPT 5 mini OpenAI | 0.0% | $0.03 | 16 |
Click row to see model details