Search results

Bug Report: ValueError with Paper Hyperparameters (training_batch_size vs ppo_mini_batch)

Status: Open.

#13 In sunnweiwei/FoldAgent;

· ji-zhe opened on Mar 27, 2026

Has anyone successfully trained Qwen3-8B with the provided training code?

Status: Open.

#12 In sunnweiwei/FoldAgent;

· hb-studying opened on Mar 5, 2026

BrowseComp reproduction with Seed-OSS-36B-Instruct via vLLM yields 0 score

Status: Open.

#11 In sunnweiwei/FoldAgent;

· hb-studying opened on Mar 5, 2026

Bug fixes and improvements for reproducibility

Status: Open.

#10 In sunnweiwei/FoldAgent;

· xzk1234 opened on Mar 4, 2026

Return type mismatch in training script. train_fold.py

Status: Open.

#9 In sunnweiwei/FoldAgent;

· rexxy-sasori opened on Feb 14, 2026

Can you share the grpo seed-oss-36b model?

Status: Open.

#8 In sunnweiwei/FoldAgent;

· tx1103mark opened on Dec 8, 2025

Inquiry for open-sourcing SWE-Bench eval

Status: Open.

#7 In sunnweiwei/FoldAgent;

· chenzhuofu opened on Nov 16, 2025

Release Folding Agent models and dataset on Hugging Face

Status: Open.

#1 In sunnweiwei/FoldAgent;

· NielsRogge opened on Oct 19, 2025