github-featured/kvcache-ai--ktransformers
2
0
Fork 0
Code Issues Pull Requests Deployments Models Agent Notes Packages Projects Releases Wiki Activity

Commit Graph

  • 95009ea685 Add MiniMax-M3 and MiniMax-M2.7 to the kt-cli model registry (#2112) main Octopus 2026-08-21 17:53:58 +08:00
  • 3464d394e1 [build]: sync sglang submodule to bc7f0058fafe6c05dc6ad52d6b1c854af2a673b2 auto/sync-sglang ovowei 2026-08-19 10:03:33 +00:00
  • 41e54fd44c [docs]: use official LlamaFactory repository in VLM guide (#2165) liyuemathematician 2026-08-18 20:10:50 +08:00
  • 517e7a4368 fix(build): keep AVX2 wheel free of AVX512 flags (#2164) yyj 2026-08-18 17:28:00 +08:00
  • 497ad7b5df fix(build): keep AVX2 wheel free of AVX512 flags fix/avx2-wheel-feature-isolation yyj 2026-08-18 17:26:13 +08:00
  • d406efc848 ci: run release jobs on self-hosted runner (#2163) yyj 2026-08-18 17:15:40 +08:00
  • 25402bb66b ci: run release jobs on self-hosted runner fix/release-self-hosted-runner yyj 2026-08-18 17:14:31 +08:00
  • ea81f4e9e5 ci: automate coordinated KT PyPI releases (#2162) yyj 2026-08-18 17:09:59 +08:00
  • a05e5165ba ci: automate coordinated KT PyPI releases release/automate-kt-sft-pypi yyj 2026-08-18 17:06:07 +08:00
  • d28b0c2eca deploy: e80614fc2d gh-pages yyj6666667 2026-08-17 08:33:06 +00:00
  • e80614fc2d [feat](kt-sft): Activation reuse & Int8 kernel refactor & native block-FP8 LoRA SFT (#2141) v0.7.0 yyj 2026-08-17 16:25:14 +08:00
  • 168f581676 [feat](sft): support Qwen VLM MoE fine-tuning (#2156) liyuemathematician 2026-08-17 10:51:55 +08:00
  • 52b46826d4 Delete kt-kernel/test/test_swiglu_limit_method_guards.py support-glm5-next-djw yyj 2026-08-15 22:00:55 +08:00
  • 7b1cd66475 feat: support SwiGLU limit for block FP8 experts yyj 2026-08-12 03:13:13 +08:00
  • eb9b70c411 [build]: sync sglang submodule (#2069) github-actions[bot] 2026-08-08 18:45:46 +08:00
  • f5ee50d4cf [docs] dsv4 tutorial update(#2140) yyj 2026-08-05 18:01:01 +08:00
  • e03cf1cefe docs: add DeepSeek V4 Flash Docker guide (#2136) yyj 2026-08-05 00:11:27 +08:00
  • 4514fa461c docs: add DeepSeek V4 Flash Docker guide docs/deepseek-v4-flash-docker yyj 2026-08-04 17:23:58 +08:00
  • 1a14f473cc [build] Sync sglang-kt submodule #2117 yyj 2026-08-02 20:11:37 +08:00
  • a8062bfa7e release: bump version to 0.6.4 (#2105) yyj 2026-07-23 22:42:28 +08:00
  • 924754a00b [feat](kt-kernel): end-to-end full-parameter and LoRA SFT(#2094) v0.6.4 yyj 2026-07-23 21:23:34 +08:00
  • d1a3ed8a30 [perf]: add K2 RAWINT4 prefill mat-mat dispatch (#2080) yyj 2026-07-19 21:43:01 +08:00
  • 937f61c988 [fix](kt-kernel): AVX-VNNI-256 RAWINT4 per-expert weight loading (#2092) Aaryan Mehta 2026-07-19 05:24:52 -04:00
  • def0f9313d fix(security): bind balance_serve scheduler ZMQ socket to loopback (#2087) (#2091) Tai An 2026-07-18 20:38:22 -07:00
  • 821f3b6198 deploy: 0c2912a5e2 yyj6666667 2026-07-17 08:24:56 +00:00
  • 0c2912a5e2 [feat] Add SYCL backend for intel iGPU (#2089) callmegaga 2026-07-17 16:24:04 +08:00
  • 01fdfa609e docs(DSV4): add supplementary deployment and tuning notes Graham 2026-07-15 00:06:58 -05:00
  • ef590428a1 deploy: 7c021b430c yyj6666667 2026-07-10 03:21:48 +00:00
  • 7c021b430c [chore](sglang): bump sglang submodule for FP16 FE8M0 Marlin fix (#2082) callmegaga 2026-07-10 11:18:45 +08:00
  • cb9f47d142 [fix](cli): detect bound ports before launch (#2071) VectorPeak 2026-07-06 18:25:31 +08:00
  • 79b265b2f6 fix: normalize compressed RAWINT4 weights (#2075) lutianshu824 2026-07-06 18:06:52 +08:00
  • 8e46e5896c release: bump version to 0.6.3.post1 (#2063) v0.6.3.post1 Benjamin 2026-06-25 22:42:55 +08:00
  • 459df61262 [build] Sync sglang-kt submodule (v0.6.3) (#2055) github-actions[bot] 2026-06-25 20:41:04 +08:00
  • a0c7431187 ci(release-pypi): make release pipeline self-consistent (#2062) Benjamin 2026-06-25 20:05:06 +08:00
  • 808b549c16 deploy: 983a88b620 yyj6666667 2026-06-24 07:23:06 +00:00
  • 7c752ebd30 deploy: 08d70e8605 yyj6666667 2026-06-24 07:22:49 +00:00
  • ecdeeb9923 deploy: 983a88b620 yyj6666667 2026-06-24 07:14:42 +00:00
  • 983a88b620 docs: add FAQ entry about SGLang MoE kernel config warning (#2029) Willow Lopez 2026-06-24 15:12:54 +08:00
  • 08d70e8605 [fix](kt-kernel): enable -mavx512vl for AVX512_VBMI/BF16 multi-variant builds (#2021) Hermann Hans Klie 2026-06-24 09:05:36 +02:00
  • 9924c9f1bb deploy: 1ed332b6ea jdai0 2026-06-23 12:52:37 +00:00
  • 1ed332b6ea [docs]: clean up Qwen3.5 KT LoRA serving guide (#2057) Jiaheng Dai 2026-06-23 20:52:23 +08:00
  • 19fc69d597 deploy: c884dbd221 yyj6666667 2026-06-21 09:55:38 +00:00
  • c884dbd221 release: bump version to 0.6.3 (#2054) Benjamin 2026-06-21 17:55:24 +08:00
  • ce7c3ddbe9 [build]: sync sglang submodule to 8b636f9008dbad58c0a8e481b03e794739e6c146 (#2047) v0.6.3 github-actions[bot] 2026-06-21 17:37:52 +08:00
  • 6c9c95601d [fix](kt-kernel): CudaGraph replay fix and Add MoE startup log (#2037) Benjamin 2026-06-18 17:15:37 +08:00
  • 943cc4daeb [feat] MXFP8 MoE support (#2041) Benjamin 2026-06-18 16:18:49 +08:00
  • 21d5b40323 docs(deepseek-v4-flash): pin apache-tvm-ffi<0.1.12 to avoid tilelang TVM FFI registration clash (#2045) (#2048) Tai An 2026-06-17 20:10:56 -07:00
  • 50d9434800 [docs](kt-kernel): add MiniMax-M3 SGLang + KT-Kernel tutorial (#2051) Benjamin 2026-06-18 10:40:51 +08:00
  • cc3d9a65bd deploy: 0f2e7905e2 ovowei 2026-06-17 06:26:21 +00:00
  • 0f2e7905e2 Update README.md (#2049) Jianwei Dong 2026-06-17 14:24:23 +08:00
  • a4fef71aff Update README.md ovowei-patch-3 Jianwei Dong 2026-06-17 14:24:11 +08:00
  • 0e2ff783d8 update glm52 tutorial (#2046) Jianwei Dong 2026-06-16 17:15:23 +08:00
  • 542d146e51 glm52 tutorial support-glm52-djw ovowei 2026-06-16 17:03:08 +08:00
  • 7641f5445d [feat] : Add MACA backend support for kt-kernel and fix CPU MoE tests (#2044) Dayuxiaoshui 2026-06-16 16:49:06 +08:00
  • 89d30a3d01 [fix(loader)]: correct off-by-one expert-count guard in load_experts (#2026) devangpratap 2026-06-07 11:41:04 -04:00
  • c1cb22311b [build]: sync sglang submodule to 51032b71279d9038058563f8d2e758d99b278ef4 (#2032) github-actions[bot] 2026-06-05 19:24:22 +08:00
  • a14b6dd5ec deploy: c9a915e6ac jdai0 2026-06-05 08:57:40 +00:00
  • c9a915e6ac [feat](kt-lora): add end-to-end Qwen3.5 MoE KT LoRA serving workflow (#2031) Jiaheng Dai 2026-06-05 16:57:14 +08:00
  • d41f569e84 [fix](cli): detect SGLANG_DSV4_2604_SUBMODE conflict before launch (#2025) devangpratap 2026-05-30 07:20:47 -04:00
  • ef6c47f9d2 [feat](kt-kernel): AVX2 MXFP4 MoE MXFP4 dispatch (#2015) Benjamin 2026-05-30 19:20:16 +08:00
  • f1e2b82c74 [fix] Add runtime AMX BF16 check to prevent SIGILL on pre-Sapphire Rapids CPUs (#2018) Li Tingfang 2026-05-21 17:36:12 +08:00
  • b77e0975da deploy: eeeeae5e91 yyj6666667 2026-05-20 07:05:04 +00:00
  • eeeeae5e91 Fix duplicate BF16 loader definition (#1984) login256 2026-05-20 15:04:47 +08:00
  • f0772445a1 [perf]: native path for MXFP4 MoE on AVX512F (#2006) Jim James 2026-05-18 02:44:33 -05:00
  • 95e20f9c55 [build]: sync sglang submodule to ebaff7729b9e41c29d94f8d19a53473d321dc566 (#2005) github-actions[bot] 2026-05-14 22:25:31 +08:00
  • f05b4009f3 [fix](kt-kernel): fix double mem used by safetensor loader (#1997) Benjamin F 2026-05-11 12:00:30 +08:00
  • bb15fdf47e Release/0.6.2.post3: carry kt-kernel SwiGLU clamp companion missing from post2 v0.6.2.post3 Benjamin F 2026-05-10 03:55:02 +08:00
  • a0f9b299bc docs: add entrypoints and support matrix doc-reorg-entrypoints-support-matrix JimmyPeilinLi 2026-05-10 01:22:13 +08:00
  • 37db9a3b83 0.6.2.post2: submodule refactor and update tutorial (#1993) v0.6.2.post2 Benjamin F 2026-05-09 18:53:59 +08:00
  • f7c4fa68c5 [fix]: add guard for SFT MoE and remove guard for AMX FP4 MoE on AVX512F+BW (#1980) Jim James 2026-05-08 03:05:22 -05:00
  • c465557c23 docs(v4-flash): add optional AMXINT4 CPU-weight conversion path (#1986) Benjamin F 2026-05-08 15:35:05 +08:00
  • 8b9d233d42 docs(v4-flash): tilelang install, MTP flags, Ampere unsupported (#1979) Benjamin F 2026-05-06 17:29:38 +08:00
  • d7b5b49a3e [release]: 0.6.2.post1 v0.6.2.post1 Benjamin F 2026-05-03 21:07:23 +08:00
  • 96189972d8 build: bump sglang submodule to c9edb75e0 (V4-Flash GPU prefill fallback fix + perf) (#1975) Benjamin F 2026-05-03 19:42:19 +08:00
  • 088ed979d5 docs(v4-flash): pin transformers==4.57.1 in tutorial prerequisites (#1974) Benjamin F 2026-05-03 16:07:31 +08:00
  • 4b4312c0a2 release: bump version to 0.6.2 (#1973) v0.6.2 Benjamin F 2026-05-03 14:28:09 +08:00
  • bb3b6e8413 build: bump sglang submodule to 40d3a82 (V4-Flash flashinfer guard) (#1972) Benjamin F 2026-05-03 14:06:33 +08:00
  • 53f356c328 deploy: 041bdfc636 yyj6666667 2026-05-03 02:48:51 +00:00
  • 041bdfc636 [New Model] DeepSeek-V4-Flash: kt-kernel MXFP4 MoE + sglang hybrid inference (#1970) Benjamin F 2026-05-03 10:48:31 +08:00
  • fe06c4d355 [build]: sync sglang submodule to 537eb762b0881071a0e098bd78666fe052b83deb (#1967) github-actions[bot] 2026-05-02 12:42:04 +08:00
  • fb4e11db95 deploy: 02be2bf53f jdai0 2026-04-30 09:17:11 +00:00
  • 02be2bf53f [feat](kt-kernel): add AVX2/AVX-VNNI RAWINT4 MoE backend (#1942) Aliez Ren 2026-04-30 18:16:49 +09:00
  • 07f39626ae deploy: 8c634d5dca JimmyPeilinLi 2026-04-30 08:25:52 +00:00
  • 8c634d5dca [docs]: refresh kt inference and sft entry points Peilin Li 2026-04-30 16:25:34 +08:00
  • 24b1941b85 [fix]: point sglang extra to post2 (#1964) v0.6.1 Peilin Li 2026-04-30 11:57:02 +08:00
  • 72044ad65f [build]: bump v0.6.1 post1 package metadata v0.6.1.post1 Peilin Li 2026-04-30 01:02:44 +08:00
  • ef5822639f [fix](kt-kernel): pin torch 2.9.1 wheel baseline Peilin Li 2026-04-30 00:57:24 +08:00
  • 9f34ef46e6 [fix](Qwen3 series): fix gibberish output by correcting RoPE write-back (#31) (#1959) Benjamin F 2026-04-27 22:04:29 +08:00
  • 1cb73c7100 [fix](Qwen3 series): fix gibberish output by correcting RoPE write-back (#31) bump-sglang-pr31 yyj 2026-04-27 21:59:38 +08:00
  • 6fea6b7d99 deploy: 0656e01ac1 JimmyPeilinLi 2026-04-26 16:46:01 +00:00
  • 0656e01ac1 [docs]: refresh KT install commands (#1958) Peilin Li 2026-04-27 00:45:43 +08:00
  • d93ea7e21e [docs]: refresh KT install commands docs-v061-refresh JimmyPeilinLi 2026-04-26 16:29:20 +00:00
  • a7a575d41e [perf](kt-kernel): MXFP4 MoE add mat-mat 4×4 tile, refine mat-vec reduce (#1957) fp4-moe-amx Benjamin F 2026-04-26 17:34:08 +08:00
  • 07e274467a [build]: flatten ktransformers package shim (#1955) Peilin Li 2026-04-25 22:08:52 +08:00
  • d143cf3209 [build]: flatten ktransformers package shim flatten-ktransformers-shim JimmyPeilinLi 2026-04-25 14:01:20 +00:00
  • bfbd0e9352 [chore]: archive kt-sft package (#1954) Peilin Li 2026-04-25 21:49:21 +08:00
  • bdf01a24b2 [chore]: archive kt-sft package archive-kt-sft JimmyPeilinLi 2026-04-25 13:42:50 +00:00
  • 85f1ab530b [ci]: use hosted runner for sglang-kt release Peilin Li 2026-04-25 21:05:18 +08:00
  • bc7afff13b [chore]: sync sglang-kt packaging fix Peilin Li 2026-04-25 21:02:25 +08:00
  • 8484ef8b16 [feat](kt-kernel): adapt MXFP4 MoE backend for DeepSeek-V4-Flash (#1950) Benjamin F 2026-04-25 18:11:53 +08:00