Visual Question Answering on EmbSpatial Bench
79AccuracyRobix-32B-Base
Evaluation Results
| Method | Links | |
|---|---|---|
| Robix-32B-BaseMethod Category=Training-Based Models, Model Scale=32B2025.11 | 79 | |
| Gemini-2.5-ProMethod Category=Generalist VLMs2025.11 | 78.4 | |
| RoboBrain2.0-32BMethod Category=Training-Based Models, Model Scale=32B2025.11 | 76.6 | |
| SandboxVLMMethod Category=Test-time Scaling VLMs2025.11 | 75.4 | |
| MindJourneyMethod Category=Test-time Scaling VLMs2025.11 | 74.7 | |
| GPT-5-miniMethod Category=Generalist VLMs2025.11 | 74.3 | |
| Qwen2.5-VL-72BMethod Category=Generalist VLMs, Model Scale=72B2025.11 | 73.3 | |
| Qwen2.5-VL-32BMethod Category=Generalist VLMs, Model Scale=32B2025.11 | 72.3 | |
| VeBrain-8BMethod Category=Training-Based Models, Model Scale=8B2025.11 | 70.5 | |
| GPT-4oMethod Category=Generalist VLMs2025.11 | 68.3 | |
| Magma-8BMethod Category=Training-Based Models, Model Scale=8B2025.11 | 64.6 | |
| Cosmos-Reason1Method Category=Training-Based Models2025.11 | 64.6 | |
| Claude-Sonnet-4Method Category=Generalist VLMs2025.11 | 64.3 |