Share your thoughts, 1 month free Claude Pro on usSee more

Embodied Navigation on EB-Navigation

85Average Score

GPT-5

Updated 7d ago

Evaluation Results

Method	Links
GPT-5 2026.05		85	85	87	85	83	85
Claude-4.6 2026.05		85	85	85	85	85	85
Qwen3-32B 2026.05		84	83	82	85	82	87
GPT-4o 2026.05		83	80	85	85	80	83
Claude-3.7 2026.05		83	85	82	83	85	82
Qwen2.5-32B 2026.05		81	75	83	83	82	83
Qwen3.5-35B 2026.05		81	80	80	80	83	82
Qwen3-8B 2026.05		80	85	85	77	75	77
Claude-4.6 2026.05		62	68	63	77	55	48
GPT-5 2026.05		57	62	62	68	50	42
Claude-3.7 2026.05		55	67	62	63	50	35
GPT-4o 2026.05		46	58	50	47	33	43
Qwen3-8B 2026.05		42	62	48	45	32	22
Qwen3-32B 2026.05		40	52	47	55	35	10
Qwen3.5-35B 2026.05		39	50	42	48	47	7
Qwen2.5-32B 2026.05		35	48	42	33	30	23