Qwen3.8-27B SWE-rebench GRPO: Step 8
This is a Hugging Face safetensors export of the completed Step 8
full-parameter GRPO checkpoint. The language-model tensors are converted from
the Miles/Megatron training checkpoint. The visual and MTP tensors, tokenizer,
configuration, and processor assets come from Qwen/Qwen3.8-27B at revision
1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0.
The run trained on 1,586 SWE-rebench v2 Filtered tasks using independent
mini-swe-agent rollouts. It used eight NVIDIA B300 GPUs (TP=2, CP=4),
32 prompt groups with eight rollouts each per optimizer step, global batch
size 256, and learning rate 2e-6. This is the no-branch GRPO baseline
checkpoint, not a branching-policy checkpoint.
Source training-checkpoint completion marker SHA-256:
006843311874c1a6fb467aaedee05ad84919eb6a82ef0613e8fd287b6ceffe18.
The original Qwen3.8-27B license is included in this repository.