mirror of
https://github.com/kvcache-ai/ktransformers.git
synced 2026-09-13 03:37:25 +00:00
[script]: add Kimi LoRA script (#2202)
release-contract-tests.yml #661 -Commit
440df94802
pushed by
vrr
ci: promote manually validated wheels without exporting PyPI credentials (#2199)
release-fake-tag.yml #658 -Commit
27b04cef5c
pushed by
vrr
ci: promote manually validated wheels without exporting PyPI credentials (#2199)
release-contract-tests.yml #657 -Commit
27b04cef5c
pushed by
vrr
ci: promote manually validated wheels without exporting PyPI credentials (#2199)
deploy.yml #656 -Commit
27b04cef5c
pushed by
vrr
ci: promote manually validated wheels without exporting PyPI credentials (#2199)
book-ci.yml #655 -Commit
27b04cef5c
pushed by
vrr
[feat] Add native RAWINT4 routed-expert LoRA SFT for Kimi K2.5/K2.6 (#2186)
deploy.yml #654 -Commit
aa2cd3072f
pushed by
vrr
[feat] Add native RAWINT4 routed-expert LoRA SFT for Kimi K2.5/K2.6 (#2186)
book-ci.yml #653 -Commit
aa2cd3072f
pushed by
vrr
[feat](docker): add docker image and API guide (#2191)
deploy.yml #652 -Commit
9b0da97c91
pushed by
vrr
[feat](docker): add docker image and API guide (#2191)
book-ci.yml #651 -Commit
9b0da97c91
pushed by
vrr
[feat](kt-kernel): packed-int4-resident AVX-VNNI-256 GPTQ INT4 backend (#2187)
sync-sglang-submodule.yml #650 -Scheduled
[feat](kt-kernel): packed-int4-resident AVX-VNNI-256 GPTQ INT4 backend (#2187)
sync-sglang-submodule.yml #649 -Scheduled
[feat](kt-kernel): packed-int4-resident AVX-VNNI-256 GPTQ INT4 backend (#2187)
deploy.yml #648 -Commit
6d460cc107
pushed by
vrr
[feat](kt-kernel): packed-int4-resident AVX-VNNI-256 GPTQ INT4 backend (#2187)
book-ci.yml #647 -Commit
6d460cc107
pushed by
vrr
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #646 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #645 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #644 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #643 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #642 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
sync-sglang-submodule.yml #641 -Scheduled
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
deploy.yml #640 -Commit
31985f40bc
pushed by
vrr
[docs](kt-kernel): DeepSeek-V4-Flash single-NPU deployment tutorial for Ascend (#2159)
book-ci.yml #639 -Commit
31985f40bc
pushed by
vrr
[perf](kt-kernel): make AVX512 MXFP4 decode memory-bound (#2181)
deploy.yml #638 -Commit
3a19fd6d88
pushed by
vrr
[perf](kt-kernel): make AVX512 MXFP4 decode memory-bound (#2181)
book-ci.yml #637 -Commit
3a19fd6d88
pushed by
vrr
[build] sync SGLang for KT LoRA serving (#2182)
release-sglang-kt.yml #636 -Commit
21095b9912
pushed by
vrr
[build] sync SGLang for KT LoRA serving (#2182)
deploy.yml #635 -Commit
21095b9912
pushed by
vrr
[build] sync SGLang for KT LoRA serving (#2182)
book-ci.yml #634 -Commit
21095b9912
pushed by
vrr
[perf](kt-kernel): AVX2 MXFP4 grouped fast path for decode and prefill (#2175)
sync-sglang-submodule.yml #633 -Scheduled