ViperEkura a03504a280 perf: preallocate inference decode buffers ...

- add InferenceWorkspace with fixed-shape per-step buffers (input_ids, decode mask, KV bind metadata) for CUDA-graph capture
- bind_tasks derives seq_lens from the pool's own _task_len tracking, dropping the seq_lens parameter
- update decode metadata in-place (position_ids, seq_lens, kv_indptr) instead of re-allocating per step
- task_extend advances _task_len in contiguous mode so the pool tracks current length
- skip log_softmax when logprobs are not requested

2026-08-03 00:55:26 +08:00

..

data

test: prune low-value and duplicate tests

2026-08-01 16:01:20 +08:00

extension

perf: preallocate inference decode buffers

2026-08-03 00:55:26 +08:00

inference

perf: preallocate inference decode buffers

2026-08-03 00:55:26 +08:00

module

feat: add MoE auxiliary loss metrics

2026-08-02 06:30:43 +08:00

optim

feat: add Mano manifold optimizer (mano_adamw)

2026-08-01 08:51:08 +08:00

parallel

fix: broadcast ref/old model state_dict for FSDP

2026-07-31 08:32:22 +08:00

trainer

refactor: emit strategy metrics as floats

2026-08-02 06:38:28 +08:00

conftest.py

fix: broadcast ref/old model state_dict for FSDP

2026-07-31 08:32:22 +08:00

helpers.py

fix: broadcast ref/old model state_dict for FSDP

2026-07-31 08:32:22 +08:00

test_train_cli.py

Make Nora+NAdamW the default optimizer

2026-07-31 23:16:39 +08:00