List your company

Embodied Agent Instruction Following on ALFWorld (ALF) (test)

91Success RateEPPO
-3.6420.9345.570.07Jul 8, 2026

Evaluation results

MethodLinks
EPPOCategory=RL Training2026.0791
AgentRLCategory=RL Training2026.0789.2
AgentLM-70BCategory=Open LLMs, Parameters=70B2026.0786
AgentLM-7BCategory=Open LLMs, Parameters=7B2026.0784
BAPOCategory=RL Training2026.0780.9
AgentLM-13BCategory=Open LLMs, Parameters=13B2026.0776
Claude-Sonnet-4Category=API LLMs2026.0773.6
Claude-Sonnet-4 ThinkingCategory=API LLMs, Variant=Thinking2026.0769
GPT-5Category=API LLMs2026.0765.4
DeepSeek-R1Category=Open LLMs2026.0751.4
Hephaestus-8B-IFTCategory=Open LLMs, Parameters=8B, Variant=IFT2026.0746
o4-miniCategory=API LLMs2026.0732.6
Qwen2.5-32B-InstructCategory=Open LLMs, Parameters=32B, Variant=Instruct2026.0732.1
DeepSeek-V3Category=Open LLMs2026.0731.9
Hephaestus-8B-BaseCategory=Open LLMs, Parameters=8B, Variant=Base2026.0730
o3-miniCategory=API LLMs2026.0728.4
Qwen2.5-14B-InstructCategory=Open LLMs, Parameters=14B, Variant=Instruct2026.078.7
Qwen2.5-3B-InstructCategory=Open LLMs, Parameters=3B, Variant=Instruct2026.072.2
DCPOCategory=RL Training2026.070