https://pypi.org/project/lk-moe/ Recent updates to the Python Package Index for lk-moe en https://pypi.org/project/lk-moe/2.0.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sat, 06 Jun 2026 23:30:09 GMT https://pypi.org/project/lk-moe/2.0.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Fri, 05 Jun 2026 09:45:37 GMT https://pypi.org/project/lk-moe/2.0.0/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Fri, 05 Jun 2026 06:29:51 GMT https://pypi.org/project/lk-moe/1.8.4/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 06 Apr 2026 12:52:20 GMT https://pypi.org/project/lk-moe/1.8.3/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 18 Mar 2026 17:19:31 GMT https://pypi.org/project/lk-moe/1.8.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Fri, 13 Mar 2026 15:24:09 GMT https://pypi.org/project/lk-moe/1.8.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 12 Mar 2026 01:17:32 GMT https://pypi.org/project/lk-moe/1.8.0/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 11 Mar 2026 00:59:33 GMT https://pypi.org/project/lk-moe/1.7.6/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 12 Feb 2026 08:03:11 GMT https://pypi.org/project/lk-moe/1.7.5/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 11 Feb 2026 16:11:48 GMT https://pypi.org/project/lk-moe/1.7.4/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 05 Feb 2026 11:42:54 GMT https://pypi.org/project/lk-moe/1.7.3/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 05 Feb 2026 04:58:07 GMT https://pypi.org/project/lk-moe/1.7.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 04 Feb 2026 08:43:38 GMT https://pypi.org/project/lk-moe/1.7.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 04 Feb 2026 04:54:24 GMT https://pypi.org/project/lk-moe/1.7.0/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 02 Feb 2026 09:12:55 GMT https://pypi.org/project/lk-moe/1.6.7/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sat, 31 Jan 2026 14:51:00 GMT https://pypi.org/project/lk-moe/1.6.6/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sat, 31 Jan 2026 09:12:13 GMT https://pypi.org/project/lk-moe/1.6.5/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 29 Jan 2026 23:58:04 GMT https://pypi.org/project/lk-moe/1.6.4/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 28 Jan 2026 01:56:36 GMT https://pypi.org/project/lk-moe/1.6.3/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Tue, 27 Jan 2026 09:58:05 GMT https://pypi.org/project/lk-moe/1.6.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Tue, 27 Jan 2026 07:31:34 GMT https://pypi.org/project/lk-moe/1.6.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 26 Jan 2026 07:51:31 GMT https://pypi.org/project/lk-moe/1.5.8/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Fri, 23 Jan 2026 17:55:11 GMT https://pypi.org/project/lk-moe/1.5.7/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Wed, 21 Jan 2026 00:07:31 GMT https://pypi.org/project/lk-moe/1.5.6/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Tue, 20 Jan 2026 22:49:41 GMT https://pypi.org/project/lk-moe/1.5.5/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 12 Jan 2026 08:01:45 GMT https://pypi.org/project/lk-moe/1.5.4/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 11 Jan 2026 16:30:00 GMT https://pypi.org/project/lk-moe/1.5.3/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sat, 10 Jan 2026 15:00:26 GMT https://pypi.org/project/lk-moe/1.5.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Thu, 08 Jan 2026 01:53:35 GMT https://pypi.org/project/lk-moe/1.4.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Tue, 06 Jan 2026 04:32:49 GMT https://pypi.org/project/lk-moe/1.4.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 04 Jan 2026 19:27:28 GMT https://pypi.org/project/lk-moe/1.4.0/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 04 Jan 2026 02:17:22 GMT https://pypi.org/project/lk-moe/1.3.5/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Tue, 30 Dec 2025 20:51:48 GMT https://pypi.org/project/lk-moe/1.3.4/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 29 Dec 2025 08:43:20 GMT https://pypi.org/project/lk-moe/1.3.3/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Mon, 29 Dec 2025 06:58:26 GMT https://pypi.org/project/lk-moe/1.3.2/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 28 Dec 2025 17:25:28 GMT https://pypi.org/project/lk-moe/1.3.1/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 28 Dec 2025 14:46:10 GMT https://pypi.org/project/lk-moe/1.3.0/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sun, 28 Dec 2025 13:27:44 GMT https://pypi.org/project/lk-moe/1.2.7/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Sat, 20 Dec 2025 16:16:15 GMT https://pypi.org/project/lk-moe/1.2.6/ lk_moe is a special NUMA extension of vllm that makes full use of CPU and memory resources, reduces GPU memory requirements, and features an efficient GPU parallel and NUMA parallel architecture, supporting hybrid inference for MOE large models. Fri, 19 Dec 2025 23:08:17 GMT