ktransformers

mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-07-14 19:17:57 +00:00

Files

devangpratap 89d30a3d01 [fix(loader)]: correct off-by-one expert-count guard in load_experts (#2026 )

* [fix(loader)]: correct off-by-one expert-count guard in SafeTensorLoader.load_experts

After the discovery loop, max_experts_count is the highest expert index found
(expert count - 1), and is -1 only when the key has no experts. The guard
checked == 0, which falsely rejected single-expert layers and silently returned
empty weight lists for the zero-expert case. Check == -1 instead.

Adds a CPU regression test covering the single-, zero-, and multi-expert cases.

* [test(loader)]: import loader as a top-level module in expert-count guard test

Per review feedback: add python/utils to sys.path and import loader directly
instead of the importlib.util boilerplate. Still bypasses utils/__init__.py
(and the compiled kt_kernel_ext) while keeping the import idiomatic.

2026-06-07 23:41:04 +08:00

cli

[fix](cli): detect SGLANG_DSV4_2604_SUBMODE conflict before launch (#2025 )

2026-05-30 19:20:47 +08:00

sft

[feat](kt-lora): add end-to-end Qwen3.5 MoE KT LoRA serving workflow (#2031 )

2026-06-05 16:57:14 +08:00

utils

[fix(loader)]: correct off-by-one expert-count guard in load_experts (#2026 )

2026-06-07 23:41:04 +08:00

__init__.py

[fix](kt-kernel): pin torch 2.9.1 wheel baseline