Progress Estimation on Real Humanoid
74.65MRAGPT-4o
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-4oAccess=Closed-Source2026.03 | 74.65 | 25.35 | |
| Claude-Haiku-4.5Access=Closed-Source2026.03 | 72.68 | 27.32 | |
| PRIMO R1Parameters=7B2026.03 | 72.32 | 21.37 | |
| Gemini 2.5 FlashAccess=Closed-Source2026.03 | 67.37 | 32.63 | |
| GPT-5 miniAccess=Closed-Source2026.03 | 67.14 | 32.59 | |
| InternVL 3.5 8BAccess=Open-Source, Parameters=8B2026.03 | 65.44 | 34.55 | |
| Qwen2.5-VL-72BAccess=Open-Source, Parameters=72B2026.03 | 62.29 | 28.1 | |
| Cosmos-Reasoning 7BType=Reasoning & Video MLLM, Parameters=7B2026.03 | 59.39 | 31.41 | |
| Qwen2.5-VL-7BAccess=Open-Source, Parameters=7B2026.03 | 56.46 | 34.73 | |
| Robobrain 7BType=Reasoning & Video MLLM, Parameters=7B2026.03 | 55.83 | 28.51 | |
| Video R1 7BType=Reasoning & Video MLLM, Parameters=7B2026.03 | 53.57 | 31.87 | |
| ProgressLM-3B-RLType=Reasoning & Video MLLM, Parameters=3B2026.03 | 32 | 34.9 |