VectorPeak
cb9f47d142
[fix](cli): detect bound ports before launch ( #2071 )
...
* [fix](cli): detect bound ports before launch
* [fix](cli): align port reuse check by platform
2026-07-06 18:25:31 +08:00
lutianshu824
79b265b2f6
fix: normalize compressed RAWINT4 weights ( #2075 )
...
* fix: normalize compressed RAWINT4 weights
* docs: add Hygon DCU ROCm notes
---------
Co-authored-by: lutianshu824 <lutianshu824@users.noreply.github.com >
2026-07-06 18:06:52 +08:00
Dayuxiaoshui
7641f5445d
[feat] : Add MACA backend support for kt-kernel and fix CPU MoE tests ( #2044 )
...
* Add MACA backend support for kt-kernel
* Add MACA event API mappings
* Fix AMX build flags and CPU MoE tests
---------
Co-authored-by: <Engle_Chaveztih@sociologist.com >
2026-06-16 16:49:06 +08:00
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
Jiaheng Dai
c9a915e6ac
[feat](kt-lora): add end-to-end Qwen3.5 MoE KT LoRA serving workflow ( #2031 )
...
* [feat](kt-lora): add KT expert LoRA adapter serving
* [feat]: pin Qwen3.5 non-expert LoRA support
* [feat](kt-lora): add merged SGLang adapter workflow
Document the KT SFT to SGLang serving loop and extend the converter with optional split outputs so users can serve one merged adapter while retaining debug-friendly expert/non-expert artifacts.
Co-authored-by: Cursor <cursoragent@cursor.com >
* [fix](kt-lora): validate adapter conversion
Co-authored-by: Cursor <cursoragent@cursor.com >
---------
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-06-05 16:57:14 +08:00
Aliez Ren
02be2bf53f
[feat](kt-kernel): add AVX2/AVX-VNNI RAWINT4 MoE backend ( #1942 )
...
* [feat](kt-kernel): add AVX2/AVX-VNNI RAWINT4 MoE backend
* Update AVX2 tutorial with AVX2 compilation instructions
Added instructions for forcing AVX2 compilation on AVX512 or AMX machines.
* Add instructions for AVX2 compilation
---------
Co-authored-by: Jiaheng Dai <108478605+jdai0@users.noreply.github.com >
2026-04-30 17:16:49 +08:00
callmegaga
a9411f1d72
Supports vnni-256 for GPTQ INT4 ( #1926 )
...
* [feat](kt-kernel): support avx-vnni-256 for gptq int4
2026-04-13 17:59:59 +08:00
mrhaoxx
7a9daf0cd4
[feat](kt-kernel): support avx2 only inference for bf16 fp8 and gptq int4 ( #1892 )
...
* feat: support avx2 bf16 fp8 inference
* feat: support avx2 gptq int4 inference
* fix: numeric issues in fp8 dequant
* Tutorial avx2 (#1900 )
* fix: prevent injecting -DLLAMA_AVX512=ON on AVX2-only machines
* docs: add AVX2 tutorial for running KTransformers on AVX2-only CPUs
* Tutorial avx2 (#1901 )
* fix: prevent injecting -DLLAMA_AVX512=ON on AVX2-only machines
* docs: add AVX2 tutorial for running KTransformers on AVX2-only CPUs
* docs: update README.md
---------
Co-authored-by: Benjamin F <159887351+yyj6666667@users.noreply.github.com >
2026-03-27 14:45:02 +08:00
Jianwei Dong
027832c590
[feat](kt-kernel): CPU-GPU experts sched ( #1796 )
2026-01-16 17:01:15 +08:00
ErvinXie
d8046e1bb4
Kt minimax ( #1742 )
...
[feat]: fp8 kernel and kt-cli support
2025-12-24 15:39:44 +08:00
Jianwei Dong
1f79f6da92
[feat](kt-kernel): Add automatic deployment workflow ( #1719 )
2025-12-16 15:20:06 +08:00
ZiWei Yuan
c2b8c60c4e
[ci]: add int4_1 & int4_1k ( #1653 )
...
* [feat]: init amd adaption
* [feat]: add blis support
* [fix]: fix setup and moe kernel warpper
* [fix](setup.py): support rebuild with cache and import kt_kernel works
fine
* [feat]: add moe_kernel converter for amd and implement the load
method(haven't tested yet)
* [feat](moe_kernel/moe.hpp): delete unused memory when using save
* [fix](moe_kernel): update PLAIN for pack
* [fix](moe_kernel): rm printf debug
* [fix](moe_kernel): skip gpu experts
* [fix](moe_kernel/moe.hpp): update include memory path
* [feat](moe_kernel/moe.hpp): support expert deferral
* [feat]: finish amd
* [ci]: add int4_1 & int4_1k
---------
Co-authored-by: mrhaoxx <mr.haoxx@gmail.com >
2025-12-02 15:58:14 +08:00
Jianwei Dong
c256150e08
update ci test ( #1647 )
2025-11-27 16:39:48 +08:00
Jianwei Dong
fef6dd98a8
add accuracy and performance test ( #1643 )
2025-11-27 10:56:39 +08:00
Jianwei Dong
51745a9ea1
add ci ( #1642 )
2025-11-25 20:52:08 +08:00
