List your company

Code Generation on HumanEval+ and LiveCodeBench

84.1Eval+ ScoreBaseline
-2.63619.88242.464.918Oct 15, 2025

Evaluation results

MethodLinks
BaselineModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=None2025.1084.143.163.6
REAPModel=Kimi-K2-Instruct-W4A16, Compression=25%2025.10844464
REAPModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=25%2025.1083.141.662.4
BaselineModel=Kimi-K2-Instruct-W4A16, Compression=None2025.1082.843.463.1
EANModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=25%2025.1082.741.962.3
REAPModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=50%2025.1082.241.561.9
REAPModel=Kimi-K2-Instruct-W4A16, Compression=50%2025.1081.942.962.4
EANModel=Kimi-K2-Instruct-W4A16, Compression=25%2025.1077.937.957.9
EANModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=50%2025.1077.738.258
FrequencyModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=25%2025.1073.729.651.6
EANModel=Kimi-K2-Instruct-W4A16, Compression=50%2025.1072.225.348.7
FrequencyModel=Kimi-K2-Instruct-W4A16, Compression=25%2025.1048.68.228.4
FrequencyModel=Kimi-K2-Instruct-W4A16, Compression=50%2025.1011.205.6
FrequencyModel=Qwen3-Coder-480B-A35B-Instruct-FP8, Compression=50%2025.100.71.21