Code Generation on LiveCodeBench V5 (Score metric) benchmark leaderboard · SOTA2 Research

Code Generation on LiveCodeBench V5 (Score metric)

80.2ScoreQwen3-235B-A22B-Thinking-2507

Evaluation Results

Method	Links
Qwen3-235B-A22B-Thinking-25072026.03		80.2
IQuest-Coder-V1-40B-Loop-Thinking2026.03		79.6
IQuest-Coder-V1-40B-Thinking2026.03		77.3
NanoV3 Elastic-30B2026.05		72.7
IQuest-Coder-V1-14B-Thinking2026.03		72.5
NanoV3-30B*2026.05		71.75
Qwen3-30B-A3B2026.05		68.25
NebulaExp-reasoning-SFT2026.06		68.25
NanoV3 Elastic-23B2026.05		67.3
Qwen3-8B-thinking2026.06		65.78
Qwen3-14B2026.06		63.5
QWQ-32B2026.06		62.7
IQuest-Coder-V1-40B-Instruct2026.03		55.7
NanoV3 Elastic-12B2026.05		55.24
Qwen3-235B-A22B-Instruct-25072026.03		53.9
Kimi-K2-Instruct-09052026.03		52.1
Qwen3-Coder-480B-A35B-Instruct2026.03		48.6
IQuest-Coder-V1-40B-Loop-Instruct2026.03		48.6
Kimi-Dev-72B2026.03		46.1
Qwen3-Coder-30B-A3B-Instruct2026.03		43.1
IQuest-Coder-V1-7B-Thinking2026.03		37.7
IQuest-Coder-V1-14B-Instruct2026.03		37.7
KAT-Dev2026.03		32.9
Qwen2.5-Coder-32B-Instruct2026.03		30.5
IQuest-Coder-V1-7B-Instruct2026.03		24.6
Qwen2.5-Coder-14B-Instruct2026.03		22.8
Seed-Coder-8B-Instruct2026.03		19.2
Qwen2.5-Coder-7B-Instruct2026.03		14.4
KAT-Dev-72B-Exp2026.03		13.8
DeepSeek-Coder-V2-Lite-Instruct2026.03		13.2