[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78Z2SSYK4TYC35V9EDZXQZ4":3,"skill-same:01M78Z2SSYK4TYC35V9EDZXQZ4":67},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M78Z2SSYK4TYC35V9EDZXQZ4","model-deployment",null,"Guides LazyLLM model modules, online providers, local serving, fine-tuning, multimodal examples, and backend-aware deployment diagnostics.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: model-deployment\ndescription: \"Guides LazyLLM model modules, online providers, local serving,\n  fine-tuning, multimodal examples, and backend-aware deployment diagnostics.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# LazyLLM Model Deployment\n\nUse this sub-skill for LazyLLM tasks about `TrainableModule`, `OnlineModule`, `OnlineChatModule`, `ServerModule`, `ActionModule`, model type mapping, model deployment, fine-tuning, distillation, local serving, provider calls, streaming tool calls, or multimodal examples.\n\n## Start here when\n\n- The user asks how to wrap a local model or online provider in LazyLLM.\n- The task mentions `lazyllm deploy`, vLLM, LMDeploy, LightLLM, Infinity, LLaMA-Factory, model caches, CUDA, or serving ports.\n- A provider\u002Fchat module fails due to model type, API key, timeout, message inspection, streaming tool calls, or provider response format.\n- The user wants to adapt chatbot, online chatbot, multimodal, painting, TTS, STT, OCR, fine-tuning, or distillation examples.\n\n## Files to read\n\n- [model-workflows.md](references\u002Fmodel-workflows.md) for class selection, online\u002Flocal\u002Fmultimodal recipes, and safe provider-free checks.\n- [backend-matrix.md](references\u002Fbackend-matrix.md) for optional dependency and backend classification.\n- [troubleshooting.md](references\u002Ftroubleshooting.md) for provider, CUDA, model path, and serving failures.\n- [scripts\u002Fmodel_surface_smoke.py](scripts\u002Fmodel_surface_smoke.py) for a no-network check of model type inference and online chat message utilities.\n\n## Safe workflow\n\n1. **Classify the module type.**\n   - `OnlineModule` \u002F `OnlineChatModule`: provider or endpoint backed.\n   - `TrainableModule`: local model\u002Ffine-tune\u002Fdeploy abstraction.\n   - `ServerModule`: wraps functions\u002Fmodules\u002Fservices and may start\u002Fbind servers.\n   - `ActionModule`: wraps deterministic actions in a LazyLLM module chain.\n2. **Classify backend requirements.** Do not run a model call until provider credentials or local model\u002FGPU\u002Fbackend are confirmed.\n3. **Perform no-network checks first.**\n   ```bash\n   python scripts\u002Fmodel_surface_smoke.py\n   python ..\u002F..\u002Fscripts\u002Finspect_lazyllm_surface.py --include-optional\n   ```\n4. **Route composed apps to their owner.** If model modules are only nodes inside a `pipeline`, read [flow-orchestration](..\u002Fflow-orchestration\u002FSKILL.md). If they serve RAG or agents, read the RAG or agent sub-skill too.\n5. **Record optional verification clearly.** Online\u002FGPU\u002Fmodel-download tests are optional unless the user explicitly asks to execute them and supplies resources.\n\n## Verified signatures to remember\n\n- `TrainableModule(base_model='', target_path='', *, stream=False, return_trace=False, trust_remote_code=True, type=None, source=None, use_model_map=True)`\n- `OnlineModule(model=None, source=None, *, type=None, url=None, **kwargs)`\n- `OnlineChatModule(model=None, source=None, url=None, stream=True, return_trace=False, skip_auth=False, type=None, api_key=None, static_params=None, id=None, name=None, group_id=None, dynamic_auth=False, timeout=180, **kwargs)`\n- `ServerModule(m=None, pre=None, post=None, stream=False, return_trace=False, port=None, pythonpath=None, launcher=None, url=None, num_replicas=1, security_key=None)`\n- `ActionModule(*action, return_trace=False)`\n\n## Model type and streaming facts\n\nLazyLLM has provider model-name mapping helpers that classify names into categories including `llm`, `vlm`, `stt`, `tts`, `embed`, `sd`, `text2video`, and `cross_modal_embed`. Tests also verify online chat helper behavior that removes prior tool traces and merges streamed tool-call chunks by index.\n\nUse these pure helpers to debug provider message formatting without calling the provider.\n\n## Backend boundaries\n\n- **Safe CPU checks:** signature inspection, model type mapping, chat history sanitization, stream chunk merge utility.\n- **Provider-required:** online chat\u002Fcompletion, online multimodal, web-search-backed flows, writer pipeline with cloud LLM.\n- **GPU\u002Flocal backend:** vLLM\u002FLMDeploy\u002FLightLLM, local TrainableModule inference, fine-tuning, distillation, painting, speech, OCR.\n- **External service:** deployed server endpoints, k8s\u002Fslurm\u002FSCO launcher tests.\n\n## Handoff outputs\n\nWhen you finish a model-deployment task, leave:\n\n- chosen LazyLLM module class and constructor parameters,\n- exact optional extra\u002Fbackend\u002Fprovider requirements,\n- safe smoke result or reason it was not executable,\n- explicit status of credentials, model path\u002Fcache, GPU, server port, and budget,\n- routes to flow\u002FRAG\u002Fagent\u002Fwriter sub-skills for composed apps.\n","# LazyLLM Model Deployment\n\nUse this sub-skill for LazyLLM tasks about `TrainableModule`, `OnlineModule`, `OnlineChatModule`, `ServerModule`, `ActionModule`, model type mapping, model deployment, fine-tuning, distillation, local serving, provider calls, streaming tool calls, or multimodal examples.\n\n## Start here when\n\n- The user asks how to wrap a local model or online provider in LazyLLM.\n- The task mentions `lazyllm deploy`, vLLM, LMDeploy, LightLLM, Infinity, LLaMA-Factory, model caches, CUDA, or serving ports.\n- A provider\u002Fchat module fails due to model type, API key, timeout, message inspection, streaming tool calls, or provider response format.\n- The user wants to adapt chatbot, online chatbot, multimodal, painting, TTS, STT, OCR, fine-tuning, or distillation examples.\n\n## Files to read\n\n- [model-workflows.md](references\u002Fmodel-workflows.md) for class selection, online\u002Flocal\u002Fmultimodal recipes, and safe provider-free checks.\n- [backend-matrix.md](references\u002Fbackend-matrix.md) for optional dependency and backend classification.\n- [troubleshooting.md](references\u002Ftroubleshooting.md) for provider, CUDA, model path, and serving failures.\n- [scripts\u002Fmodel_surface_smoke.py](scripts\u002Fmodel_surface_smoke.py) for a no-network check of model type inference and online chat message utilities.\n\n## Safe workflow\n\n1. **Classify the module type.**\n   - `OnlineModule` \u002F `OnlineChatModule`: provider or endpoint backed.\n   - `TrainableModule`: local model\u002Ffine-tune\u002Fdeploy abstraction.\n   - `ServerModule`: wraps functions\u002Fmodules\u002Fservices and may start\u002Fbind servers.\n   - `ActionModule`: wraps deterministic actions in a LazyLLM module chain.\n2. **Classify backend requirements.** Do not run a model call until provider credentials or local model\u002FGPU\u002Fbackend are confirmed.\n3. **Perform no-network checks first.**\n   ```bash\n   python scripts\u002Fmodel_surface_smoke.py\n   python ..\u002F..\u002Fscripts\u002Finspect_lazyllm_surface.py --include-optional\n   ```\n4. **Route composed apps to their owner.** If model modules are only nodes inside a `pipeline`, read [flow-orchestration](..\u002Fflow-orchestration\u002FSKILL.md). If they serve RAG or agents, read the RAG or agent sub-skill too.\n5. **Record optional verification clearly.** Online\u002FGPU\u002Fmodel-download tests are optional unless the user explicitly asks to execute them and supplies resources.\n\n## Verified signatures to remember\n\n- `TrainableModule(base_model='', target_path='', *, stream=False, return_trace=False, trust_remote_code=True, type=None, source=None, use_model_map=True)`\n- `OnlineModule(model=None, source=None, *, type=None, url=None, **kwargs)`\n- `OnlineChatModule(model=None, source=None, url=None, stream=True, return_trace=False, skip_auth=False, type=None, api_key=None, static_params=None, id=None, name=None, group_id=None, dynamic_auth=False, timeout=180, **kwargs)`\n- `ServerModule(m=None, pre=None, post=None, stream=False, return_trace=False, port=None, pythonpath=None, launcher=None, url=None, num_replicas=1, security_key=None)`\n- `ActionModule(*action, return_trace=False)`\n\n## Model type and streaming facts\n\nLazyLLM has provider model-name mapping helpers that classify names into categories including `llm`, `vlm`, `stt`, `tts`, `embed`, `sd`, `text2video`, and `cross_modal_embed`. Tests also verify online chat helper behavior that removes prior tool traces and merges streamed tool-call chunks by index.\n\nUse these pure helpers to debug provider message formatting without calling the provider.\n\n## Backend boundaries\n\n- **Safe CPU checks:** signature inspection, model type mapping, chat history sanitization, stream chunk merge utility.\n- **Provider-required:** online chat\u002Fcompletion, online multimodal, web-search-backed flows, writer pipeline with cloud LLM.\n- **GPU\u002Flocal backend:** vLLM\u002FLMDeploy\u002FLightLLM, local TrainableModule inference, fine-tuning, distillation, painting, speech, OCR.\n- **External service:** deployed server endpoints, k8s\u002Fslurm\u002FSCO launcher tests.\n\n## Handoff outputs\n\nWhen you finish a model-deployment task, leave:\n\n- chosen LazyLLM module class and constructor parameters,\n- exact optional extra\u002Fbackend\u002Fprovider requirements,\n- safe smoke result or reason it was not executable,\n- explicit status of credentials, model path\u002Fcache, GPU, server port, and budget,\n- routes to flow\u002FRAG\u002Fagent\u002Fwriter sub-skills for composed apps.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill model-deployment",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill model-deployment",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill model-deployment","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002FSKILL.md",[47,53,57,61],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fbackend-matrix.md","REFERENCE","参考文档",2738,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002Freferences\u002Fbackend-matrix.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fmodel-workflows.md",3775,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002Freferences\u002Fmodel-workflows.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Ftroubleshooting.md",2489,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002Freferences\u002Ftroubleshooting.md",{"path":62,"resourceType":63,"resourceTypeDesc":64,"sizeBytes":65,"sourceHtmlUrl":66},"scripts\u002Fmodel_surface_smoke.py","SCRIPT","脚本",2820,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Flazy-llm\u002Fsub-skills\u002Fmodel-deployment\u002Fscripts\u002Fmodel_surface_smoke.py",[68,75,82,89,96,103],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":104,"name":105,"nameZh":6,"description":106,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]