List your company

Code Generation on LiveCodeBench v6 Aug 2024-May 2025 (test)

63.1Pass@1FADE h* = 1.0
13.28426.21739.1552.083Jul 1, 2026

Evaluation results

MethodLinks
FADE h* = 1.0Backbone=CWM 32B, Budget=30k2026.0763.177.985.386.3
Power α asym (2,2)Backbone=CWM 32B, Budget=30k2026.0762.876.283.384.6
GRPOBackbone=CWM 32B, Budget=30k2026.076276.783.785
Power α = 2Backbone=CWM 32B, Budget=30k2026.0761.276.782.983.9
FADE h* = 1.0Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0740.75868.2—
Power α = 2Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0740.356.666—
Power α asym (2,2)Backbone=Qwen 2.5 7B SFT, Budget=8k2026.07405664.9—
Implicit p@2Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0739.756.265.2—
Asym. normBackbone=Qwen 2.5 7B SFT, Budget=8k2026.0738.153.463—
Implicit p@8Backbone=Qwen 2.5 7B SFT, Budget=8k2026.073854.364.7—
SFT modelBackbone=CWM 32B, Budget=30k2026.0737.5546972.5
GRPOBackbone=Qwen 2.5 7B SFT, Budget=8k2026.0737.453.662.5—
LME β = 0.4Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0735.952.562—
Asym δ = 0.5Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0735.751.860.5—
Pass@8Backbone=Qwen 2.5 7B SFT, Budget=8k2026.0733.851.761.1—
REINFORCEBackbone=Qwen 2.5 7B SFT, Budget=8k2026.0729.445.355.8—
SFT modelBackbone=Qwen 2.5 7B SFT, Budget=8k2026.0715.242.352.2—