vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-04-28 11:49:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

Fix kt-kernel for new wrapper (#1588)
Some checks are pending
Book-CI / test (push) Waiting to run
Details
Book-CI / test-1 (push) Waiting to run
Details
Book-CI / test-2 (push) Waiting to run
Details
Deploy / deploy (macos-latest) (push) Waiting to run
Details
Deploy / deploy (ubuntu-latest) (push) Waiting to run
Details
Deploy / deploy (windows-latest) (push) Waiting to run
Details

Browse source 
* update README for kt-kernel

* style: format C++ and Python code in kt-kernel

  - Format C++ files: task_queue, ext_bindings, and MoE operators
  - Format Python utility modules: amx, llamafile, and loader
  - Improve code readability and consistency
This commit is contained in:
Jiaqi Liao 2025-11-10 21:47:34 +08:00 • committed by GitHub
parent 9bc00e587b
commit 94c25626dc
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 219 additions and 179 deletions
Download patch file Download diff file Expand all files Collapse all files

11
kt-kernel/python/__init__.py
View file

@ -6,8 +6,8 @@ KT-Kernel provides high-performance kernel operations for KTransformers,
including CPU-optimized MoE inference with AMX, AVX, and KML support.
Example usage:
>>> from kt_kernel import AMXMoEWrapper
>>> wrapper = AMXMoEWrapper(
>>> from kt_kernel import KTMoEWrapper
>>> wrapper = KTMoEWrapper(
... layer_idx=0,
... num_experts=8,
... num_experts_per_tok=2,
@ -15,9 +15,10 @@ Example usage:
... moe_intermediate_size=14336,
... num_gpu_experts=2,
... cpuinfer_threads=32,
... subpool_count=2,
... amx_weight_path="/path/to/weights",
... chunked_prefill_size=512
... threadpool_count=2,
... weight_path="/path/to/weights",
... chunked_prefill_size=512,
... method="AMXINT4"
... )
"""