Repository navigation

All pull requests

Add the single-device Qwen3-TTS LoRA cookbook
#384Wallbreazzz opened Nov 12, 2026
Add the Qwen3-TTS dual-channel input processor
#383Wallbreazzz opened Nov 12, 2026
Register the Qwen3-TTS residual codebook loss
#382Wallbreazzz opened Nov 12, 2026
Add the Qwen3-TTS two-stage training forward patch
#381Wallbreazzz opened Nov 12, 2026
Defer pretrained architecture resolution to the model loader
#380Wallbreazzz opened Nov 12, 2026
fix(transformers): correct FSDP LoRA dtype initialization and MoE SP router alignment
#379werwrewe opened Nov 12, 2026Contributor1 of 4 tasks
3
Adaptive mopd weight
#377indigoRM opened Nov 12, 20264 tasks
fix(auto): detect the accelerator before starting the training server
#371li-lizhe opened Nov 7, 2026Contributor1 of 4 tasks
feat(model): move DP balancing into TrainableModel, enabled by default
#367lincq2000 opened Nov 3, 2026Collaborator2 of 4 tasks
RL续训
#340indigoRM opened Oct 26, 20261 of 4 tasks
fix(loss): compute the IPO loss as in Eq. 17 of the IPO paper
#327flyingdoubleG opened Oct 21, 2026Contributor2 of 4 tasks
1
fix(loss): truncate the CISPO IS weight at 1 + epsilon_high
#326flyingdoubleG opened Oct 21, 2026Contributor2 of 4 tasks
fix: restore local optimizer state without distributed broadcast
#317Liuchenbing-2026 opened Oct 14, 2026Contributor1 of 4 tasks
1
fix: update on complete gradient accumulation boundaries
#316Liuchenbing-2026 opened Oct 14, 2026Contributor1 of 4 tasks
1
feat: add native typed decision training components
#314Liuchenbing-2026 opened Oct 14, 2026Contributor3 of 4 tasks
feat: add StartLux decision LoRA training on Ascend
#305Liuchenbing-2026 opened Oct 12, 2026Contributor4 tasks done
feat: add Intern-Decision text training cookbook on Ascend
#295Liuchenbing-2026 opened Oct 8, 2026Contributor2 of 3 tasks
fix: weight PPO token-mean micro-batches by valid tokens
#292taking-lying-flat opened Sep 20, 20261 of 4 tasks
fix: detach probability weights in DFT loss
#291taking-lying-flat opened Sep 20, 20261 of 4 tasks
fix: avoid duplicate gradient norm reduction in pure DDP
#290taking-lying-flat opened Sep 20, 20261 of 4 tasks
[feature]reward_loop
#284doctorMcy opened Sep 15, 20263 tasks
feat(rl): add SAO synchronous training baseline with skip-observation GAE, DIS policy loss and value critic ```
#269xxyyrr598 opened Aug 19, 2026Contributor
6
add java client
#268ljhandhya opened Aug 18, 20263 tasks
Fix/multi lora ep memory efficient init
#265meichangsu1 opened Aug 14, 2026Collaborator4 tasks
[New feature] 支持多教师在线蒸馏（MOPD）
#239doctorMcy opened Jun 30, 20263 tasks
10

Pull requests