List your company

Visual Planning on Maze (Path Metrics)

7.2Path Length Metric 3LLaVA-OneVision-7B
4.423.342.261.1Jun 4, 2026

Evaluation results

MethodLinks
LLaVA-OneVision-7BInput Modality=Visual, Prompting Strategy=Prompt-only2026.067.245.64.45.3
InternVL3-8BInput Modality=Visual, Prompting Strategy=Prompt-only2026.06107.67.647.3
Qwen2.5-VL-3B-InstructInput Modality=Visual, Prompting Strategy=Prompt-only2026.06105.65.65.26.6
Qwen2.5-VL-7B-InstructInput Modality=Visual, Prompting Strategy=Prompt-only2026.0611.2610.83.67.9
Qwen3-VL-4B-InstructInput Modality=Visual, Prompting Strategy=Prompt-only2026.0624.4125.24.811.6
Qwen3-VL-235B-A22B-InstructInput Modality=Visual, Prompting Strategy=Prompt-only2026.06322422.416.423.7
Claude-4.5-HaikuInput Modality=Visual, Prompting Strategy=Prompt-only2026.0634.83021.615.225.4
Qwen3-VL-8B-InstructInput Modality=Visual, Prompting Strategy=Prompt-only2026.0635.22015.28.819.8
GPT-4oInput Modality=Visual, Prompting Strategy=Prompt-only2026.063830.420.416.826.4
Gemini-2.5-ProInput Modality=Visual, Prompting Strategy=Prompt-only2026.0641.221.613.68.821.3
Gemini-2.5-FlashInput Modality=Visual, Prompting Strategy=Prompt-only2026.064432.822.815.628.8
MGSD-4BInput Modality=Visual, Task-specific training status=MGSD Trained2026.0648.635.621.614.829.7
Symbolic-Input Qwen3-VL-4BInput Modality=Symbolic2026.0650.438.425.617.232.9
MGSD-8BInput Modality=Visual, Task-specific training status=MGSD Trained2026.06523628.41833.6
Qwen3-VL-32B-ThinkingInput Modality=Visual, Prompting Strategy=Prompt-only2026.0654.43419.217.631.3
Kimi-K2.5Input Modality=Visual, Prompting Strategy=Prompt-only2026.0658.443.633.620.439
Symbolic-Input Qwen3-VL-8BInput Modality=Symbolic2026.0661.64631.629.642.2
Gemini-3-FlashInput Modality=Visual, Prompting Strategy=Prompt-only2026.0675.654.43016.444.1
GPT-5Input Modality=Visual, Prompting Strategy=Prompt-only2026.0677.244.426.41641