List your company

Code Generation on MBPP (tau, Speed, Score)

91.39Overall ScoreCoSpec
67.792473.918780.04586.1713May 24, 2026May 30, 2026Jun 5, 2026Jun 11, 2026Jun 17, 2026Jun 23, 2026Jun 29, 2026

Evaluation results

MethodLinks
CoSpecSetting=T = 0, Target=LLaMA-70B2026.0591.393.9220.96
CoSpecSetting=T = 1, Target=LLaMA-70B2026.0590.113.9721.13
AASDSetting=T = 0, Target=LLaMA-70B2026.0590.031.722.08
AASDSetting=T = 1, Target=LLaMA-70B2026.0589.41.752.16
Target onlySetting=T = 0, Target=LLaMA-70B2026.0589.1811
SpSSetting=T = 0, Target=LLaMA-70B2026.0589.181.6810.43
RESTSetting=T = 0, Target=LLaMA-70B2026.0589.182.012.34
EAGLE-3Setting=T = 0, Target=LLaMA-70B2026.0589.183.775.45
Target onlySetting=T = 1, Target=LLaMA-70B2026.0588.7311
SpSSetting=T = 1, Target=LLaMA-70B2026.0588.731.516.08
RESTSetting=T = 1, Target=LLaMA-70B2026.0588.731.942.24
EAGLE-3Setting=T = 1, Target=LLaMA-70B2026.0588.733.585.15
FLySetting=T = 0, Target=LLaMA-70B2026.0588.622.4511.76
FLySetting=T = 1, Target=LLaMA-70B2026.0588.422.6912.33
ScoreFlow2026.0684.7——
AFlow2026.0682.9——
CoSpecSetting=T = 1, Target=Qwen3-32B2026.0580.091.9617.97
CoSpecSetting=T = 0, Target=Qwen3-32B2026.0579.11.9718.01
MetaFlowgeneration_mode=single-inference generation2026.0677.5——
Target onlySetting=T = 1, Target=Qwen3-32B2026.0576.1111
SpSSetting=T = 1, Target=Qwen3-32B2026.0576.111.279.41
EAGLE-3Setting=T = 1, Target=Qwen3-32B2026.0576.111.833.8
Target onlySetting=T = 0, Target=Qwen3-32B2026.0575.9211
SpSSetting=T = 0, Target=Qwen3-32B2026.0575.921.339.52
EAGLE-3Setting=T = 0, Target=Qwen3-32B2026.0575.921.913.89
AASDSetting=T = 1, Target=Qwen3-32B2026.0575.721.582
AASDSetting=T = 0, Target=Qwen3-32B2026.0575.61.592.03
FLySetting=T = 0, Target=Qwen3-32B2026.0575.421.5810.55
FLySetting=T = 1, Target=Qwen3-32B2026.0575.381.5810.12
CoT SC2026.0671.3——
CoT2026.0670.4——
MultiPersona2026.0670.4——
Self Refine2026.0670——
IO2026.0669.5——
MedPrompt2026.0669.2——
ADAS2026.0668.7——