Our new X account is live! Follow @wizwand_team for updates

Code Reasoning on MBPP

84.7MBPP Execution Accuracy

Qwen2.5-Coder-CFB-Aug

Updated 4d ago

Evaluation Results

Method	Links
Qwen2.5-Coder-CFB-Aug 2024.10		84.7	69.3
Qwen2.5-Coder-ScaleQuest 2024.10		83.1	69.9
Qwen2.5-Coder-CFB 2024.10		77.2	64
Denser 2025.12		73.8	-
Process Supervision 2025.12		73.2	-
Reflection-CoT 2025.12		72.6	-
Self-Verification 2025.12		72.6	-
RePro 2025.12		72.1	-
Tree-of-Thought 2025.12		71.9	-
Denser 2025.12		71.7	-
Process Supervision 2025.12		71.6	-
Self-Consistency 2025.12		71.2	-
Reflection-CoT 2025.12		71	-
Think-to-Think 2025.12		70.5	-
Self-Verification 2025.12		70.4	-
Tree-of-Thought 2025.12		70.3	-
Original 2025.12		70	-
Self-Consistency 2025.12		69.6	-
Process Supervision 2025.12		69.5	-
Denser 2025.12		69.5	-
Reflection-CoT 2025.12		68.9	-
Think-to-Think 2025.12		68.9	-
GRPO 2025.12		68.8	-
Chain-of-Thought 2025.12		68.5	-
Self-Verification 2025.12		68.3	-
Tree-of-Thought 2025.12		67.8	-
Self-Consistency 2025.12		67.1	-
Chain-of-Thought 2025.12		66.9	-
Think-to-Think 2025.12		66.4	-
Chain-of-Thought 2025.12		64.7	-
RePro 2025.12		59.3	-
GRPO 2025.12		58.7	-
Original 2025.12		25	-