Object classification on TEA (test)
74.48AccuracyHuman
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Human2026.02 | 74.48 | — | |
| o3-20252026.02 | 41.96 | — | |
| GPT-52026.02 | 40.56 | — | |
| GPT-4.12026.02 | 35.66 | — | |
| Gemini-2.5-Pro2026.02 | 33.57 | — | |
| Gemini-2.5-Flash2026.02 | 32.17 | — | |
| llama-4-Scout2026.02 | 30.07 | — | |
| GPT-4o2026.02 | 7.69 | — | |
| Claude-Opus-42026.02 | 4.9 | — |