Code Generation on LiveCodeBench (Score)

73.1Score

Deepseek-R1

Updated 1mo ago

Evaluation Results

Method	Links
Deepseek-R1 2026.05		73.1
SFT-based Classification Router 2026.05		70.5
MoMA Router 2026.05		66.5
Qwen3-235B-A22B 2026.05		65.9
Contrastive learning based Router 2026.05		61.3
Qwen3-32B 2026.05		60.7
DreamReasoner-8B 2026.06		53.9
DreamReasoner-8B 2026.06		53.6
Qwen3-8B-Thinking 2026.06		52.8
AceReason-Nemotron-1.1-7B 2026.06		52.1
DreamReasoner-8B 2026.06		51.3
MiMo-7B-RL 2026.06		50.7
DreamReasoner-8B 2026.06		50.4
MoMA Router 2026.05		45.3
LLaDA-2.0-Flash (100B-A6B) 2026.06		41
Contrastive learning based Router 2026.05		40.1
DeepSeek-R1-Distill-Qwen-7B 2026.06		37.8
SDAR-30B-A3B-Sci 2026.06		29
Contrastive learning based Router 2026.05		27.6
Deepseek-V3 2026.05		27.2
JT-Code-8B 2026.05		26.3
MoMA Router 2026.05		24.6
SDAR-8B-Chat 2026.06		16.4
TraDo-8B-Instruct 2026.06		16.1
TraDo-8B-Thinking 2026.06		14.2
Dream-7B-Instruct 2026.06		9.7
SDAR-30B-A3B-Sci 2026.06		6.4
SDAR-30B-A3B-Sci 2026.06		3.2
SDAR-30B-A3B-Sci 2026.06		2.6