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

..

__init__.py

Kt minimax (#1742)

2025-12-24 15:39:44 +08:00

amx.py

[feat](kt-kernel): AVX2 MXFP4 MoE MXFP4 dispatch (#2015)

2026-05-30 19:20:16 +08:00

llamafile.py

[fix]: fix --numa-nodes handling (#1904)

2026-03-31 17:50:22 +08:00

loader.py

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

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

moe_kernel.py

[fix]: fix --numa-nodes handling (#1904)

2026-03-31 17:50:22 +08:00