AstrAI

Files

T

ViperEkura 3067a8e1a6 feat: unify attention backend with multi-dim mask support

- Add attention() functional entry delegating to active backend
- GQA/MLA forward calls attention() instead of inline cache/SDPA
- CUDA kernels support 2D/3D/4D mask via mask_h_stride field
- CudaBackend.fwd_decode builds 2D padding mask for mixed seq_lens
- KVCache.max_len precomputed in bind_tasks to avoid GPU sync
- batch==1 decode short-circuits mask=None
- Split tests into conftest, test_backend, test_backend_equivalence, test_kernel_mask
- 440 tests pass, L20 decode 1.44-1.60x speedup vs torch native

2026-07-30 20:38:34 +08:00

config

feat: add field and model validators to config classes

2026-07-30 08:41:14 +08:00

dataset

refactor: remove H5 storage backend in favor of mmap bin

2026-07-29 12:50:27 +08:00

extension

feat: unify attention backend with multi-dim mask support

2026-07-30 20:38:34 +08:00

inference

feat: unify attention backend with multi-dim mask support