List your company

Task planning on MomaGraph-Bench Overall 1.0

73.9AccuracyClaude-4.5-Sonnet
37.60447.02756.4565.873Dec 18, 2025

Evaluation results

MethodLinks
Claude-4.5-SonnetReasoning Protocol=w/ Graph2025.1273.9
GPT-5Reasoning Protocol=w/ Graph2025.1271.6
Gemini-2.5-ProReasoning Protocol=w/ Graph2025.1271.6
MomaGraph-R1Params=7B, Reasoning Protocol=w/ Graph2025.1271.6
Claude-4.5-SonnetReasoning Protocol=w/o Graph2025.1269.8
Gemini-2.5-ProReasoning Protocol=w/o Graph2025.1267.6
GPT-5Reasoning Protocol=w/o Graph2025.1266.5
MomaGraph-R1Params=7B, Reasoning Protocol=w/o Graph2025.1265.1
Qwen2.5-VL-7B-InstructParams=7B, Reasoning Protocol=w/ Graph2025.1260.2
Qwen2.5-VL-7B-InstructParams=7B, Reasoning Protocol=w/o Graph2025.1257.2
LLaVA-OnevisionParams=7B, Reasoning Protocol=w/ Graph2025.1255.6
DeepSeek-VL2Params=4.5B, Reasoning Protocol=w/ Graph2025.1254.3
LLaVA-OnevisionParams=7B, Reasoning Protocol=w/o Graph2025.1253.5
DeepSeek-VL2Params=4.5B, Reasoning Protocol=w/o Graph2025.1252
InternVL2.5Params=8B, Reasoning Protocol=w/ Graph2025.1251.1
InternVL2.5Params=8B, Reasoning Protocol=w/o Graph2025.1248.4
LLaVA-V1.5Params=7B, Reasoning Protocol=w/ Graph2025.1244.8
LLaVA-V1.5Params=7B, Reasoning Protocol=w/o Graph2025.1244.1
InstructBLIPParams=7B, Reasoning Protocol=w/ Graph2025.1239.5
InstructBLIPParams=7B, Reasoning Protocol=w/o Graph2025.1239