Robot parameter extraction and forward kinematics calculation on Benchmark 1 (test)
97M_C (Completeness/Score)GPT-4o
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| GPT-4oVersion=Framework2026.02 | 97 | 100 | 100 | 100 | 100 | 97 | — | 99 | |
| DeepSeek-V3.2Version=Framework2026.02 | 93 | 93 | 100 | 93 | 93 | 93 | — | 94 | |
| DeepSeek-V3.2Version=Raw2026.02 | 80 | — | — | — | — | — | — | — | |
| Claude-Sonnet-4.5Version=Framework2026.02 | 77 | 100 | 77 | 77 | 77 | 100 | — | 86 | |
| Claude-Sonnet-4.5Version=Raw2026.02 | 40 | — | — | — | — | — | — | — | |
| GPT-4oVersion=Raw2026.02 | 30 | — | — | — | — | — | — | — |