ViperEkura a29bdfae46 refactor: rework attention backend resolution ...

- explicit attn_backend() context wins over ASTR_BACKEND env
- polymorphic available()/supports_call() replace isinstance dispatch
- cache singleton backend instances to avoid hot-path allocation
- training (fwd=None) resolves cuda > flash > torch by capability
- flash dense supports mask-free calls only; masked training falls back to torch

2026-08-23 14:47:02 +08:00

..

distributed.md

docs: audit non-CUDA documentation

2026-08-02 07:39:24 +08:00

evaluation.md

docs: sync all documentation with current codebase

2026-08-07 23:52:27 +08:00

inference.md

refactor: rework attention backend resolution

2026-08-23 14:47:02 +08:00

params.md

docs: fix stale docs and align with code

2026-08-23 14:23:28 +08:00

preprocessing.md

docs: audit non-CUDA documentation

2026-08-02 07:39:24 +08:00

training.md

docs: align CUDA kernel and RoPE docs with code

2026-08-03 20:54:40 +08:00