Planning on PlanBench

99.1Accuracy

DeepSeek-R1 Distill

Updated 1mo ago

Evaluation Results

Method	Links
DeepSeek-R1 Distill 2026.06		99.1
TeamTR (Heterogeneous) 2026.05		44.5
TeamTR (Homogeneous) 2026.05		44.1
Qwen3-Base (DICE-PC) 2026.06		42.7
Qwen3-235B-A22B 2026.06		42.5
OpenAI o3 2026.06		42.1
Qwen3-Base (DICE-FT) 2026.06		41.9
o4-mini 2026.05		41.2
OpenAI o3-mini 2026.06		41.2
Qwen3 (thinking) 2026.05		40.3
Qwen3 (thinking) 2026.06		40.3
Qwen3-Inst (Few-shot CoT) 2026.06		40.1
Qwen3-Inst (SC@64) 2026.06		39.7
Qwen3-A3B 2026.05		39.1
DeepSeek-R1 Distill 2026.05		39.1
Debate + Judge 2026.05		39.1
Qwen3-Base Debate+Judge (FT) 2026.06		39.1
Qwen3-A3B 2026.06		39.1
Role-play 2026.05		38.7
Qwen3-Base Role-play (FT) 2026.06		38.7
Qwen3-Base Debate+Judge (FT) 2026.06		38.4
Debate 2026.05		38.3
Qwen3-Base Debate (FT) 2026.06		38.3
Qwen3-Inst Debate+Judge (prompt) 2026.06		38.1
QwQ 2026.05		37.9
Qwen3-Inst (ReAct) 2026.06		37.9
Qwen3-Base Role-play (FT) 2026.06		37.9
QwQ 2026.06		37.9
Qwen3-Inst Role-play (prompt) 2026.06		37.8
Qwen3-Base Debate (FT) 2026.06		37.5
Qwen3-Inst Debate (prompt) 2026.06		37.2
Llama 3.1-Base (DICE-PC) 2026.06		37.1
Qwen3-Inst (ToT) 2026.06		37
Qwen3-Inst (SC@64) 2026.06		36.3
Qwen3-Base-DAPO 2026.06		35.6
Qwen3-Base-GRPO 2026.06		34.8
GPT-4o-mini 2026.05		34.7
GPT-4o-mini 2026.06		34.7
Qwen3-Base 2026.06		34.3
Qwen2.5-Base 2026.06		33.7
DAPO 2026.05		33.5
Qwen3-Base-DAPO 2026.06		33.5
GRPO 2026.05		33.1
Qwen3-Base-GRPO 2026.06		33.1
Qwen3-Base-PPO 2026.06		32.6
Qwen3-Base 2026.06		29.7
Qwen3-Inst (Zero-shot CoT) 2026.06		29.5