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

General Reasoning on AutoLogi

90.5Accuracy

TeamTR (Homogeneous)

Updated 2mo ago

Evaluation Results

Method	Links
TeamTR (Homogeneous) 2026.05		90.5
Qwen3-Base SAT 2026.04		89.1
TeamTR (Heterogeneous) 2026.05		88.7
Qwen3-A3B 2026.04		88.1
Qwen3-A3B 2026.05		88.1
Qwen3 (thinking) 2026.04		87.3
Qwen3 (thinking) 2026.05		87.3
QwQ 2026.04		86.3
OpenAI o3-mini (medium) 2026.04		86.3
o4-mini 2026.05		86.3
QwQ 2026.05		86.3
Qwen3-Base Debate + Judge 2026.04		85.9
Debate + Judge 2026.05		85.9
Qwen3-Base Role-play 2026.04		85.5
Role-play 2026.05		85.5
Qwen3-Base Debate 2026.04		85.1
Debate 2026.05		85.1
Llama 3.1-Base SAT 2026.04		83.9
DeepSeek-R1 Distill Llama 2026.04		83.5
DeepSeek-R1 Distill 2026.05		83.5
Qwen3-Base-DAPO 2026.04		80.9
DAPO 2026.05		80.9
Qwen3-Base-GRPO 2026.04		80.3
GRPO 2026.05		80.3
Qwen3-Base 2026.04		79.1
Qwen3-Base 2026.04		78.5
Qwen2.5-Base 2026.04		76.7
GPT-4o-mini-2024-07-18 2026.04		62.5
GPT-4o-mini 2026.05		62.5