[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6G6CYRZZM39C5SW2KMSPFJW":3,"skill-same:01M6G6CYRZZM39C5SW2KMSPFJW":62},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":33,"linkOnly":34,"markdownContent":35,"markdownBody":36,"frontMatter":37,"refName":38,"commitSha":39,"skillDirectory":40,"forksCount":41,"repositoryHtmlUrl":42,"repositoryHomepageUrl":43,"ownerHtmlUrl":44,"ownerBlogUrl":45,"ownerDescription":46,"updatedAt":47,"install":48,"resources":61},"01M6G6CYRZZM39C5SW2KMSPFJW","vllm",null,"Use this skill when the user wants to serve a local LLM behind an OpenAI-compatible API or benchmark model throughput\u002Flatency with vLLM.","javimosch\u002Fsupercli","javimosch","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F12900508?v=4","01M6D3F2AS6PV75ET9PZ1QGP8J","Javier Leandro Arancibia","01M6D3F2EBNWNJ358Q90P5YQR1",57,"MIT","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fvllm\u002Fskills\u002Fquickstart\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32],"vLLM","benchmark","openai","api","model","user","use","local llm","local","latency","serve","throughput","with","openai-compatible","2026-11-10 10:12:30",false,"---\nname: vllm\ndescription: Use this skill when the user wants to serve a local LLM behind an OpenAI-compatible API or benchmark model throughput\u002Flatency with vLLM.\n---\n\n# vllm Plugin\n\nvLLM is the reference high-throughput LLM inference\u002Fserving engine (PagedAttention, continuous batching, tensor parallelism). Its CLI serves models behind an OpenAI-compatible API and includes chat\u002Fcomplete clients and benchmarks.\n\n## Commands\n\n### Serving\n- `vllm model serve \u003Cmodel>` — Start an OpenAI-compatible server (long-running; pass `--port`, `--dtype`, `--gpu-memory-utilization`, etc.)\n- `vllm model chat` — Interactive chat client against a running server\n- `vllm model complete` — One-shot completion client\n\n### Benchmarks\n- `vllm bench throughput` — Offline throughput benchmark\n- `vllm bench latency` — Latency benchmark\n\n### Diagnostics\n- `vllm env collect` — Environment report for bug reports\n- `vllm _ _` — Passthrough (`run-batch`, `bench serve`, ...)\n\n## Usage Examples\n\n- \"Serve Qwen2.5-1.5B as an OpenAI API on port 8000\"\n- \"Benchmark throughput of this model with 256-token inputs\"\n- \"Collect my environment for a vLLM bug report\"\n\n## Installation\n\n```bash\npip install vllm      # or: uv pip install vllm\n```\n\nCUDA GPU strongly recommended; ~2GB install (bundles PyTorch).\n\n## Examples\n\n```bash\n# Serve with quantization + memory tuning\nvllm serve Qwen\u002FQwen2.5-1.5B-Instruct --port 8000 --dtype auto --gpu-memory-utilization 0.85\n\n# One-shot client calls\nvllm chat --url http:\u002F\u002Flocalhost:8000\nvllm complete --url http:\u002F\u002Flocalhost:8000\n\n# Benchmarks\nvllm bench throughput --model Qwen\u002FQwen2.5-1.5B-Instruct --input-len 256 --output-len 128\nvllm bench latency --model Qwen\u002FQwen2.5-1.5B-Instruct --batch-size 8\n```\n\n## Key Features\n\n- OpenAI-compatible `\u002Fv1\u002Fchat\u002Fcompletions`, `\u002Fv1\u002Fcompletions`, `\u002Fv1\u002Fembeddings`\n- PagedAttention KV-cache management → high throughput, low latency\n- Continuous batching + tensor\u002Fpipeline parallelism across GPUs\n- Built-in `bench` suite (throughput, latency, and `bench serve` for online load)\n","# vllm Plugin\n\nvLLM is the reference high-throughput LLM inference\u002Fserving engine (PagedAttention, continuous batching, tensor parallelism). Its CLI serves models behind an OpenAI-compatible API and includes chat\u002Fcomplete clients and benchmarks.\n\n## Commands\n\n### Serving\n- `vllm model serve \u003Cmodel>` — Start an OpenAI-compatible server (long-running; pass `--port`, `--dtype`, `--gpu-memory-utilization`, etc.)\n- `vllm model chat` — Interactive chat client against a running server\n- `vllm model complete` — One-shot completion client\n\n### Benchmarks\n- `vllm bench throughput` — Offline throughput benchmark\n- `vllm bench latency` — Latency benchmark\n\n### Diagnostics\n- `vllm env collect` — Environment report for bug reports\n- `vllm _ _` — Passthrough (`run-batch`, `bench serve`, ...)\n\n## Usage Examples\n\n- \"Serve Qwen2.5-1.5B as an OpenAI API on port 8000\"\n- \"Benchmark throughput of this model with 256-token inputs\"\n- \"Collect my environment for a vLLM bug report\"\n\n## Installation\n\n```bash\npip install vllm      # or: uv pip install vllm\n```\n\nCUDA GPU strongly recommended; ~2GB install (bundles PyTorch).\n\n## Examples\n\n```bash\n# Serve with quantization + memory tuning\nvllm serve Qwen\u002FQwen2.5-1.5B-Instruct --port 8000 --dtype auto --gpu-memory-utilization 0.85\n\n# One-shot client calls\nvllm chat --url http:\u002F\u002Flocalhost:8000\nvllm complete --url http:\u002F\u002Flocalhost:8000\n\n# Benchmarks\nvllm bench throughput --model Qwen\u002FQwen2.5-1.5B-Instruct --input-len 256 --output-len 128\nvllm bench latency --model Qwen\u002FQwen2.5-1.5B-Instruct --batch-size 8\n```\n\n## Key Features\n\n- OpenAI-compatible `\u002Fv1\u002Fchat\u002Fcompletions`, `\u002Fv1\u002Fcompletions`, `\u002Fv1\u002Fembeddings`\n- PagedAttention KV-cache management → high throughput, low latency\n- Continuous batching + tensor\u002Fpipeline parallelism across GPUs\n- Built-in `bench` suite (throughput, latency, and `bench serve` for online load)\n",{"name":5,"description":7},"master","18c32f4b7197d246bdedcc9eca5d0403eaac0bc2","plugins\u002Fvllm\u002Fskills\u002Fquickstart",5,"https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli","https:\u002F\u002Fjavimosch.github.io\u002Fsupercli\u002F","https:\u002F\u002Fgithub.com\u002Fjavimosch","intrane.fr","Building robust, open-source infrastructure for the commons. AI-leveraged, self-hosted, built to last.","2026-11-02 18:24:17",{"source":8,"skillName":5,"commands":49,"prompt":59,"zipAvailable":60},[50,53,56],{"tool":51,"command":52},"npx","npx skills add javimosch\u002Fsupercli --skill vllm",{"tool":54,"command":55},"bunx","bunx skills add javimosch\u002Fsupercli --skill vllm",{"tool":57,"command":58},"pnpm","pnpm dlx skills add javimosch\u002Fsupercli --skill vllm","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fvllm\u002Fskills\u002Fquickstart\u002FSKILL.md",true,[],[63,76,96,107,119,127],{"skillNo":64,"name":65,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":73,"publishedAt":33},"01M6G7YFZVHRYAY51J0YG9SDNJ","sc-machin-mcp","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fsupercli-machin-cli\u002Fskills\u002Fsc-machin-mcp\u002FSKILL.md",[68],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[74,75],"sc machin mcp","MCP",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":83,"publishedAt":33},"01M6G7YB179MQMSHZBJDC8TZA8","supercli.quickstart","SuperCLI capability router — discover and execute CLI commands through namespace.resource.action interface","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fskills\u002Fsupercli\u002FSKILL.md",[82],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},[84,85,86,87,88,89,90,91,92,93,94,95],"supercli quickstart","quickstart","cli","capability router","plugin","discovery","execute","router","discover","commands","interface","supercli,cli,plugin,capability,discovery,agent",{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":33},"01M6G7Y734QS3793YMS5PR146J","zzdoc","Use this skill to compile scdoc manpages with a fast Zig-based compiler.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fzzdoc\u002Fskills\u002Fquickstart\u002FSKILL.md",[],[98,31,25,103,104,105,106],"zig","fast","compiler","compile",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":33},"01M6G7Y2WX56K2QJQ9RV2ZKEV7","zx","Use this skill when the user wants to a tool for writing better scripts using node.js.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fzx\u002Fskills\u002Fquickstart\u002FSKILL.md",[],[109,114,25,24,115,116,117,118],"scripts","node","tool","js","writing",{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":33},"01M6G7Y0M792DGW31HN4TEQT10","zvm","Use this skill when the user wants to manage multiple Zig compiler versions.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fzvm\u002Fskills\u002Fquickstart\u002FSKILL.md",[],[121,103,126,25,24,105],"manage",{"skillNo":128,"name":85,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":33},"01M6G7XX51W5HJW09DYXBK92TJ","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fsupercli\u002Fblob\u002Fmaster\u002Fplugins\u002Fzstdmt\u002Fskills\u002Fquickstart\u002FSKILL.md",[],[85]]