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

Code Problem Solving on LiveCodeBench v6

100Success Rate (Easy)

MemP

Updated 1mo ago

Evaluation Results

Method	Links
MemP 2026.06		100	42.3	22.5	47.4
MemP 2026.06		100	67.3	47.5	66.3
Qwen3.5-4B 2026.06		100	84.6	55	74.9
Qwen3.5-4B 2026.06		100	91.9	64.5	81.4
SkillComposer-4B 2026.06		100	94.2	66.8	83.1
No Skill 2026.06		99.5	90.4	62	79.7
SkillComposer-4B 2026.06		99.4	93.3	67.2	82.9
Qwen3.5-4B 2026.06		99.2	90.4	64.2	80.6
No Skill 2026.06		99.1	67.3	26.8	56.6
SkillComposer-4B 2026.06		98.5	69.2	31.3	59
SkillComposer-4B 2026.06		98.3	89.4	59.7	78
Qwen3.5-4B 2026.06		98.1	68.9	26.5	56.7
SkillComposer-4B 2026.06		98.1	72.7	29.3	59.1
Qwen3.5-4B 2026.06		97.7	61.5	25	53.7
RiVER 2026.06		97.7	59	17.9	49.7
Raw-GRPO 2026.06		96.9	58.3	17.9	49.3
Instance-Norm 2026.06		96.9	56.4	19.2	49.3
RiVER 2026.06		96.9	60.3	21.2	51.4
Raw-Binary 2026.06		96.9	57.1	18.3	49.1
Rank-uniform 2026.06		96.9	53.9	18.3	48.2
Qwen3-8B 2026.06		96.3	55.6	19.7	49.2
RS-GRPO 2026.06		96.1	59.6	20	50.5
Raw-Binary 2026.06		96.1	55.1	19.6	49
RS-GRPO 2026.06		96.1	59	17.9	49.3
GLM-Z1-9B-0414 2026.06		95.6	53.3	16.1	46.7
Qwen3.5-4B 2026.06		95.4	64.1	29.6	56
SkillComposer-4B 2026.06		95.4	65.4	25.8	54.7
Rank-uniform 2026.06		95.4	59.6	17.5	49.1
Raw-GRPO 2026.06		95.3	52.6	19.6	48
Instance-Norm 2026.06		94.6	55.1	17.1	47.4