List your company

Narrow finetuning on Auto

83.4EM-F1TReFT
52.260.368.476.5Jun 4, 2026

Evaluation results

MethodLinks
TReFTModel=Qwen-2.5-32B2026.0683.4-0.2
TReFTModel=Llama-3.1-8B2026.0677.90.4
TReFTModel=Qwen-2.5-7B2026.0676.51.6
Data interleavingModel=Qwen-2.5-32B2026.0674.9-1
TReFTModel=GPT-OSS-20B2026.0674.3-0.81
Data interleavingModel=Qwen-2.5-7B2026.0670.90.3
Data interleavingModel=Llama-3.1-8B2026.0670.20.2
SFTModel=Qwen-2.5-7B2026.0668.60.6
SFTModel=GPT-OSS-20B2026.0664-2.77
Data interleavingModel=GPT-OSS-20B2026.0663.6-1.46
SFTModel=Llama-3.1-8B2026.0654.2-1
SFTModel=Qwen-2.5-32B2026.0653.4-2