Backup Qwen3.5-397B-data custom SFT adapter qwen35-397b-data-qwen35-9b-custom-sft-20260630T042600Z

#4
BenchFlow org

Open backup PR for the env-0 train-mini custom PyTorch/PEFT SFT replay. Do not merge until the behavioral eval result is reviewed.

BenchFlow org

Merging PR #4 to promote the Qwen3.5-397B-data custom SFT LoRA adapter as the current model version.

bingran-you changed pull request status to merged
BenchFlow org

Behavioral eval has been reviewed: Mobile300 134-135/300 strict pass and standard60 Fireworks 25/180 strict pass. Promoting this adapter to main.

bingran-you changed pull request status to open

Sign up or log in to comment