[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71TB2RFASMJ50QM14C0D9K5":3,"skill-same:01M71TB2RFASMJ50QM14C0D9K5":101},{"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":27,"publishedAt":48,"linkOnly":49,"markdownContent":50,"markdownBody":51,"frontMatter":52,"refName":63,"commitSha":64,"skillDirectory":65,"forksCount":13,"repositoryHtmlUrl":66,"repositoryHomepageUrl":6,"ownerHtmlUrl":67,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":68,"install":69,"resources":82},"01M71TB2RFASMJ50QM14C0D9K5","serving-llms-vllm",null,"vLLM: high-throughput LLM serving, OpenAI API, quantization.","eliottbusiness\u002FDeptFlow-Agent","eliottbusiness","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F263418316?v=4","01M71T5T608RHFA07W6QVC6TYX","01M71T5T6R98RMYZ53PYH851T1",0,"MIT","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002FSKILL.md",[17,22],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47],"llms","serving","serving llms vllm","vLLM","Production","high-throughput","Tensor Parallelism","Quantization","Transformers","Continuous Batching","PagedAttention","high","tensor","Inference Serving","research","High Throughput","llm serving","api","batching","orchestra","2026-11-10 13:02:02",false,"---\nname: serving-llms-vllm\ndescription: \"vLLM: high-throughput LLM serving, OpenAI API, quantization.\"\nversion: 1.0.0\nauthor: Orchestra Research\nlicense: MIT\ndependencies: [vllm, torch, transformers]\nmetadata:\n  hermes:\n    tags: [vLLM, Inference Serving, PagedAttention, Continuous Batching, High Throughput, Production, OpenAI API, Quantization, Tensor Parallelism]\n\n---\n\n# vLLM - High-Performance LLM Serving\n\n## When to use\n\nUse when deploying production LLM APIs, optimizing inference latency\u002Fthroughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ\u002FAWQ\u002FFP8), and tensor parallelism.\n\n## Quick start\n\nvLLM achieves 24x higher throughput than standard transformers through PagedAttention (block-based KV cache) and continuous batching (mixing prefill\u002Fdecode requests).\n\n**Installation**:\n```bash\npip install vllm\n```\n\n**Basic offline inference**:\n```python\nfrom vllm import LLM, SamplingParams\n\nllm = LLM(model=\"meta-llama\u002FLlama-3-8B-Instruct\")\nsampling = SamplingParams(temperature=0.7, max_tokens=256)\n\noutputs = llm.generate([\"Explain quantum computing\"], sampling)\nprint(outputs[0].outputs[0].text)\n```\n\n**OpenAI-compatible server**:\n```bash\nvllm serve meta-llama\u002FLlama-3-8B-Instruct\n\n# Query with OpenAI SDK\npython -c \"\nfrom openai import OpenAI\nclient = OpenAI(base_url='http:\u002F\u002Flocalhost:8000\u002Fv1', api_key='EMPTY')\nprint(client.chat.completions.create(\n    model='meta-llama\u002FLlama-3-8B-Instruct',\n    messages=[{'role': 'user', 'content': 'Hello!'}]\n).choices[0].message.content)\n\"\n```\n\n## Common workflows\n\n### Workflow 1: Production API deployment\n\nCopy this checklist and track progress:\n\n```\nDeployment Progress:\n- [ ] Step 1: Configure server settings\n- [ ] Step 2: Test with limited traffic\n- [ ] Step 3: Enable monitoring\n- [ ] Step 4: Deploy to production\n- [ ] Step 5: Verify performance metrics\n```\n\n**Step 1: Configure server settings**\n\nChoose configuration based on your model size:\n\n```bash\n# For 7B-13B models on single GPU\nvllm serve meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --max-model-len 8192 \\\n  --port 8000\n\n# For 30B-70B models with tensor parallelism\nvllm serve meta-llama\u002FLlama-2-70b-hf \\\n  --tensor-parallel-size 4 \\\n  --gpu-memory-utilization 0.9 \\\n  --quantization awq \\\n  --port 8000\n\n# For production with caching and metrics\nvllm serve meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --enable-prefix-caching \\\n  --enable-metrics \\\n  --metrics-port 9090 \\\n  --port 8000 \\\n  --host 0.0.0.0\n```\n\n**Step 2: Test with limited traffic**\n\nRun load test before production:\n\n```bash\n# Install load testing tool\npip install locust\n\n# Create test_load.py with sample requests\n# Run: locust -f test_load.py --host http:\u002F\u002Flocalhost:8000\n```\n\nVerify TTFT (time to first token) \u003C 500ms and throughput > 100 req\u002Fsec.\n\n**Step 3: Enable monitoring**\n\nvLLM exposes Prometheus metrics on port 9090:\n\n```bash\ncurl http:\u002F\u002Flocalhost:9090\u002Fmetrics | grep vllm\n```\n\nKey metrics to monitor:\n- `vllm:time_to_first_token_seconds` - Latency\n- `vllm:num_requests_running` - Active requests\n- `vllm:gpu_cache_usage_perc` - KV cache utilization\n\n**Step 4: Deploy to production**\n\nUse Docker for consistent deployment:\n\n```bash\n# Run vLLM in Docker\ndocker run --gpus all -p 8000:8000 \\\n  vllm\u002Fvllm-openai:latest \\\n  --model meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --enable-prefix-caching\n```\n\n**Step 5: Verify performance metrics**\n\nCheck that deployment meets targets:\n- TTFT \u003C 500ms (for short prompts)\n- Throughput > target req\u002Fsec\n- GPU utilization > 80%\n- No OOM errors in logs\n\n### Workflow 2: Offline batch inference\n\nFor processing large datasets without server overhead.\n\nCopy this checklist:\n\n```\nBatch Processing:\n- [ ] Step 1: Prepare input data\n- [ ] Step 2: Configure LLM engine\n- [ ] Step 3: Run batch inference\n- [ ] Step 4: Process results\n```\n\n**Step 1: Prepare input data**\n\n```python\n# Load prompts from file\nprompts = []\nwith open(\"prompts.txt\") as f:\n    prompts = [line.strip() for line in f]\n\nprint(f\"Loaded {len(prompts)} prompts\")\n```\n\n**Step 2: Configure LLM engine**\n\n```python\nfrom vllm import LLM, SamplingParams\n\nllm = LLM(\n    model=\"meta-llama\u002FLlama-3-8B-Instruct\",\n    tensor_parallel_size=2,  # Use 2 GPUs\n    gpu_memory_utilization=0.9,\n    max_model_len=4096\n)\n\nsampling = SamplingParams(\n    temperature=0.7,\n    top_p=0.95,\n    max_tokens=512,\n    stop=[\"\u003C\u002Fs>\", \"\\n\\n\"]\n)\n```\n\n**Step 3: Run batch inference**\n\nvLLM automatically batches requests for efficiency:\n\n```python\n# Process all prompts in one call\noutputs = llm.generate(prompts, sampling)\n\n# vLLM handles batching internally\n# No need to manually chunk prompts\n```\n\n**Step 4: Process results**\n\n```python\n# Extract generated text\nresults = []\nfor output in outputs:\n    prompt = output.prompt\n    generated = output.outputs[0].text\n    results.append({\n        \"prompt\": prompt,\n        \"generated\": generated,\n        \"tokens\": len(output.outputs[0].token_ids)\n    })\n\n# Save to file\nimport json\nwith open(\"results.jsonl\", \"w\") as f:\n    for result in results:\n        f.write(json.dumps(result) + \"\\n\")\n\nprint(f\"Processed {len(results)} prompts\")\n```\n\n### Workflow 3: Quantized model serving\n\nFit large models in limited GPU memory.\n\n```\nQuantization Setup:\n- [ ] Step 1: Choose quantization method\n- [ ] Step 2: Find or create quantized model\n- [ ] Step 3: Launch with quantization flag\n- [ ] Step 4: Verify accuracy\n```\n\n**Step 1: Choose quantization method**\n\n- **AWQ**: Best for 70B models, minimal accuracy loss\n- **GPTQ**: Wide model support, good compression\n- **FP8**: Fastest on H100 GPUs\n\n**Step 2: Find or create quantized model**\n\nUse pre-quantized models from HuggingFace:\n\n```bash\n# Search for AWQ models\n# Example: TheBloke\u002FLlama-2-70B-AWQ\n```\n\n**Step 3: Launch with quantization flag**\n\n```bash\n# Using pre-quantized model\nvllm serve TheBloke\u002FLlama-2-70B-AWQ \\\n  --quantization awq \\\n  --tensor-parallel-size 1 \\\n  --gpu-memory-utilization 0.95\n\n# Results: 70B model in ~40GB VRAM\n```\n\n**Step 4: Verify accuracy**\n\nTest outputs match expected quality:\n\n```python\n# Compare quantized vs non-quantized responses\n# Verify task-specific performance unchanged\n```\n\n## When to use vs alternatives\n\n**Use vLLM when:**\n- Deploying production LLM APIs (100+ req\u002Fsec)\n- Serving OpenAI-compatible endpoints\n- Limited GPU memory but need large models\n- Multi-user applications (chatbots, assistants)\n- Need low latency with high throughput\n\n**Use alternatives instead:**\n- **llama.cpp**: CPU\u002Fedge inference, single-user\n- **HuggingFace transformers**: Research, prototyping, one-off generation\n- **TensorRT-LLM**: NVIDIA-only, need absolute maximum performance\n- **Text-Generation-Inference**: Already in HuggingFace ecosystem\n\n## Common issues\n\n**Issue: Out of memory during model loading**\n\nReduce memory usage:\n```bash\nvllm serve MODEL \\\n  --gpu-memory-utilization 0.7 \\\n  --max-model-len 4096\n```\n\nOr use quantization:\n```bash\nvllm serve MODEL --quantization awq\n```\n\n**Issue: Slow first token (TTFT > 1 second)**\n\nEnable prefix caching for repeated prompts:\n```bash\nvllm serve MODEL --enable-prefix-caching\n```\n\nFor long prompts, enable chunked prefill:\n```bash\nvllm serve MODEL --enable-chunked-prefill\n```\n\n**Issue: Model not found error**\n\nUse `--trust-remote-code` for custom models:\n```bash\nvllm serve MODEL --trust-remote-code\n```\n\n**Issue: Low throughput (\u003C50 req\u002Fsec)**\n\nIncrease concurrent sequences:\n```bash\nvllm serve MODEL --max-num-seqs 512\n```\n\nCheck GPU utilization with `nvidia-smi` - should be >80%.\n\n**Issue: Inference slower than expected**\n\nVerify tensor parallelism uses power of 2 GPUs:\n```bash\nvllm serve MODEL --tensor-parallel-size 4  # Not 3\n```\n\nEnable speculative decoding for faster generation:\n```bash\nvllm serve MODEL --speculative-model DRAFT_MODEL\n```\n\n## Advanced topics\n\n**Server deployment patterns**: See [references\u002Fserver-deployment.md](references\u002Fserver-deployment.md) for Docker, Kubernetes, and load balancing configurations.\n\n**Performance optimization**: See [references\u002Foptimization.md](references\u002Foptimization.md) for PagedAttention tuning, continuous batching details, and benchmark results.\n\n**Quantization guide**: See [references\u002Fquantization.md](references\u002Fquantization.md) for AWQ\u002FGPTQ\u002FFP8 setup, model preparation, and accuracy comparisons.\n\n**Troubleshooting**: See [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for detailed error messages, debugging steps, and performance diagnostics.\n\n## Hardware requirements\n\n- **Small models (7B-13B)**: 1x A10 (24GB) or A100 (40GB)\n- **Medium models (30B-40B)**: 2x A100 (40GB) with tensor parallelism\n- **Large models (70B+)**: 4x A100 (40GB) or 2x A100 (80GB), use AWQ\u002FGPTQ\n\nSupported platforms: NVIDIA (primary), AMD ROCm, Intel GPUs, TPUs\n\n## Resources\n\n- Official docs: https:\u002F\u002Fdocs.vllm.ai\n- GitHub: https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm\n- Paper: \"Efficient Memory Management for Large Language Model Serving with PagedAttention\" (SOSP 2023)\n- Community: https:\u002F\u002Fdiscuss.vllm.ai\n\n\n\n","# vLLM - High-Performance LLM Serving\n\n## When to use\n\nUse when deploying production LLM APIs, optimizing inference latency\u002Fthroughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ\u002FAWQ\u002FFP8), and tensor parallelism.\n\n## Quick start\n\nvLLM achieves 24x higher throughput than standard transformers through PagedAttention (block-based KV cache) and continuous batching (mixing prefill\u002Fdecode requests).\n\n**Installation**:\n```bash\npip install vllm\n```\n\n**Basic offline inference**:\n```python\nfrom vllm import LLM, SamplingParams\n\nllm = LLM(model=\"meta-llama\u002FLlama-3-8B-Instruct\")\nsampling = SamplingParams(temperature=0.7, max_tokens=256)\n\noutputs = llm.generate([\"Explain quantum computing\"], sampling)\nprint(outputs[0].outputs[0].text)\n```\n\n**OpenAI-compatible server**:\n```bash\nvllm serve meta-llama\u002FLlama-3-8B-Instruct\n\n# Query with OpenAI SDK\npython -c \"\nfrom openai import OpenAI\nclient = OpenAI(base_url='http:\u002F\u002Flocalhost:8000\u002Fv1', api_key='EMPTY')\nprint(client.chat.completions.create(\n    model='meta-llama\u002FLlama-3-8B-Instruct',\n    messages=[{'role': 'user', 'content': 'Hello!'}]\n).choices[0].message.content)\n\"\n```\n\n## Common workflows\n\n### Workflow 1: Production API deployment\n\nCopy this checklist and track progress:\n\n```\nDeployment Progress:\n- [ ] Step 1: Configure server settings\n- [ ] Step 2: Test with limited traffic\n- [ ] Step 3: Enable monitoring\n- [ ] Step 4: Deploy to production\n- [ ] Step 5: Verify performance metrics\n```\n\n**Step 1: Configure server settings**\n\nChoose configuration based on your model size:\n\n```bash\n# For 7B-13B models on single GPU\nvllm serve meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --max-model-len 8192 \\\n  --port 8000\n\n# For 30B-70B models with tensor parallelism\nvllm serve meta-llama\u002FLlama-2-70b-hf \\\n  --tensor-parallel-size 4 \\\n  --gpu-memory-utilization 0.9 \\\n  --quantization awq \\\n  --port 8000\n\n# For production with caching and metrics\nvllm serve meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --enable-prefix-caching \\\n  --enable-metrics \\\n  --metrics-port 9090 \\\n  --port 8000 \\\n  --host 0.0.0.0\n```\n\n**Step 2: Test with limited traffic**\n\nRun load test before production:\n\n```bash\n# Install load testing tool\npip install locust\n\n# Create test_load.py with sample requests\n# Run: locust -f test_load.py --host http:\u002F\u002Flocalhost:8000\n```\n\nVerify TTFT (time to first token) \u003C 500ms and throughput > 100 req\u002Fsec.\n\n**Step 3: Enable monitoring**\n\nvLLM exposes Prometheus metrics on port 9090:\n\n```bash\ncurl http:\u002F\u002Flocalhost:9090\u002Fmetrics | grep vllm\n```\n\nKey metrics to monitor:\n- `vllm:time_to_first_token_seconds` - Latency\n- `vllm:num_requests_running` - Active requests\n- `vllm:gpu_cache_usage_perc` - KV cache utilization\n\n**Step 4: Deploy to production**\n\nUse Docker for consistent deployment:\n\n```bash\n# Run vLLM in Docker\ndocker run --gpus all -p 8000:8000 \\\n  vllm\u002Fvllm-openai:latest \\\n  --model meta-llama\u002FLlama-3-8B-Instruct \\\n  --gpu-memory-utilization 0.9 \\\n  --enable-prefix-caching\n```\n\n**Step 5: Verify performance metrics**\n\nCheck that deployment meets targets:\n- TTFT \u003C 500ms (for short prompts)\n- Throughput > target req\u002Fsec\n- GPU utilization > 80%\n- No OOM errors in logs\n\n### Workflow 2: Offline batch inference\n\nFor processing large datasets without server overhead.\n\nCopy this checklist:\n\n```\nBatch Processing:\n- [ ] Step 1: Prepare input data\n- [ ] Step 2: Configure LLM engine\n- [ ] Step 3: Run batch inference\n- [ ] Step 4: Process results\n```\n\n**Step 1: Prepare input data**\n\n```python\n# Load prompts from file\nprompts = []\nwith open(\"prompts.txt\") as f:\n    prompts = [line.strip() for line in f]\n\nprint(f\"Loaded {len(prompts)} prompts\")\n```\n\n**Step 2: Configure LLM engine**\n\n```python\nfrom vllm import LLM, SamplingParams\n\nllm = LLM(\n    model=\"meta-llama\u002FLlama-3-8B-Instruct\",\n    tensor_parallel_size=2,  # Use 2 GPUs\n    gpu_memory_utilization=0.9,\n    max_model_len=4096\n)\n\nsampling = SamplingParams(\n    temperature=0.7,\n    top_p=0.95,\n    max_tokens=512,\n    stop=[\"\u003C\u002Fs>\", \"\\n\\n\"]\n)\n```\n\n**Step 3: Run batch inference**\n\nvLLM automatically batches requests for efficiency:\n\n```python\n# Process all prompts in one call\noutputs = llm.generate(prompts, sampling)\n\n# vLLM handles batching internally\n# No need to manually chunk prompts\n```\n\n**Step 4: Process results**\n\n```python\n# Extract generated text\nresults = []\nfor output in outputs:\n    prompt = output.prompt\n    generated = output.outputs[0].text\n    results.append({\n        \"prompt\": prompt,\n        \"generated\": generated,\n        \"tokens\": len(output.outputs[0].token_ids)\n    })\n\n# Save to file\nimport json\nwith open(\"results.jsonl\", \"w\") as f:\n    for result in results:\n        f.write(json.dumps(result) + \"\\n\")\n\nprint(f\"Processed {len(results)} prompts\")\n```\n\n### Workflow 3: Quantized model serving\n\nFit large models in limited GPU memory.\n\n```\nQuantization Setup:\n- [ ] Step 1: Choose quantization method\n- [ ] Step 2: Find or create quantized model\n- [ ] Step 3: Launch with quantization flag\n- [ ] Step 4: Verify accuracy\n```\n\n**Step 1: Choose quantization method**\n\n- **AWQ**: Best for 70B models, minimal accuracy loss\n- **GPTQ**: Wide model support, good compression\n- **FP8**: Fastest on H100 GPUs\n\n**Step 2: Find or create quantized model**\n\nUse pre-quantized models from HuggingFace:\n\n```bash\n# Search for AWQ models\n# Example: TheBloke\u002FLlama-2-70B-AWQ\n```\n\n**Step 3: Launch with quantization flag**\n\n```bash\n# Using pre-quantized model\nvllm serve TheBloke\u002FLlama-2-70B-AWQ \\\n  --quantization awq \\\n  --tensor-parallel-size 1 \\\n  --gpu-memory-utilization 0.95\n\n# Results: 70B model in ~40GB VRAM\n```\n\n**Step 4: Verify accuracy**\n\nTest outputs match expected quality:\n\n```python\n# Compare quantized vs non-quantized responses\n# Verify task-specific performance unchanged\n```\n\n## When to use vs alternatives\n\n**Use vLLM when:**\n- Deploying production LLM APIs (100+ req\u002Fsec)\n- Serving OpenAI-compatible endpoints\n- Limited GPU memory but need large models\n- Multi-user applications (chatbots, assistants)\n- Need low latency with high throughput\n\n**Use alternatives instead:**\n- **llama.cpp**: CPU\u002Fedge inference, single-user\n- **HuggingFace transformers**: Research, prototyping, one-off generation\n- **TensorRT-LLM**: NVIDIA-only, need absolute maximum performance\n- **Text-Generation-Inference**: Already in HuggingFace ecosystem\n\n## Common issues\n\n**Issue: Out of memory during model loading**\n\nReduce memory usage:\n```bash\nvllm serve MODEL \\\n  --gpu-memory-utilization 0.7 \\\n  --max-model-len 4096\n```\n\nOr use quantization:\n```bash\nvllm serve MODEL --quantization awq\n```\n\n**Issue: Slow first token (TTFT > 1 second)**\n\nEnable prefix caching for repeated prompts:\n```bash\nvllm serve MODEL --enable-prefix-caching\n```\n\nFor long prompts, enable chunked prefill:\n```bash\nvllm serve MODEL --enable-chunked-prefill\n```\n\n**Issue: Model not found error**\n\nUse `--trust-remote-code` for custom models:\n```bash\nvllm serve MODEL --trust-remote-code\n```\n\n**Issue: Low throughput (\u003C50 req\u002Fsec)**\n\nIncrease concurrent sequences:\n```bash\nvllm serve MODEL --max-num-seqs 512\n```\n\nCheck GPU utilization with `nvidia-smi` - should be >80%.\n\n**Issue: Inference slower than expected**\n\nVerify tensor parallelism uses power of 2 GPUs:\n```bash\nvllm serve MODEL --tensor-parallel-size 4  # Not 3\n```\n\nEnable speculative decoding for faster generation:\n```bash\nvllm serve MODEL --speculative-model DRAFT_MODEL\n```\n\n## Advanced topics\n\n**Server deployment patterns**: See [references\u002Fserver-deployment.md](references\u002Fserver-deployment.md) for Docker, Kubernetes, and load balancing configurations.\n\n**Performance optimization**: See [references\u002Foptimization.md](references\u002Foptimization.md) for PagedAttention tuning, continuous batching details, and benchmark results.\n\n**Quantization guide**: See [references\u002Fquantization.md](references\u002Fquantization.md) for AWQ\u002FGPTQ\u002FFP8 setup, model preparation, and accuracy comparisons.\n\n**Troubleshooting**: See [references\u002Ftroubleshooting.md](references\u002Ftroubleshooting.md) for detailed error messages, debugging steps, and performance diagnostics.\n\n## Hardware requirements\n\n- **Small models (7B-13B)**: 1x A10 (24GB) or A100 (40GB)\n- **Medium models (30B-40B)**: 2x A100 (40GB) with tensor parallelism\n- **Large models (70B+)**: 4x A100 (40GB) or 2x A100 (80GB), use AWQ\u002FGPTQ\n\nSupported platforms: NVIDIA (primary), AMD ROCm, Intel GPUs, TPUs\n\n## Resources\n\n- Official docs: https:\u002F\u002Fdocs.vllm.ai\n- GitHub: https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm\n- Paper: \"Efficient Memory Management for Large Language Model Serving with PagedAttention\" (SOSP 2023)\n- Community: https:\u002F\u002Fdiscuss.vllm.ai\n\n\n\n",{"name":5,"author":53,"license":14,"version":54,"metadata":55,"description":7,"dependencies":59},"Orchestra Research","1.0.0",{"hermes":56},{"tags":57},[31,41,38,37,43,32,58,35,34],"OpenAI API",[60,61,62],"vllm","torch","transformers","main","3ffb682eba417bd920665a7146fecc206b3bccf8","profile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent","https:\u002F\u002Fgithub.com\u002Feliottbusiness","2026-05-03 19:19:17",{"source":8,"skillName":5,"commands":70,"prompt":80,"zipAvailable":81},[71,74,77],{"tool":72,"command":73},"npx","npx skills add eliottbusiness\u002FDeptFlow-Agent --skill serving-llms-vllm",{"tool":75,"command":76},"bunx","bunx skills add eliottbusiness\u002FDeptFlow-Agent --skill serving-llms-vllm",{"tool":78,"command":79},"pnpm","pnpm dlx skills add eliottbusiness\u002FDeptFlow-Agent --skill serving-llms-vllm","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002FSKILL.md",true,[83,89,93,97],{"path":84,"resourceType":85,"resourceTypeDesc":86,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Foptimization.md","REFERENCE","参考文档",5769,"https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002Freferences\u002Foptimization.md",{"path":90,"resourceType":85,"resourceTypeDesc":86,"sizeBytes":91,"sourceHtmlUrl":92},"references\u002Fquantization.md",6746,"https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002Freferences\u002Fquantization.md",{"path":94,"resourceType":85,"resourceTypeDesc":86,"sizeBytes":95,"sourceHtmlUrl":96},"references\u002Fserver-deployment.md",5252,"https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002Freferences\u002Fserver-deployment.md",{"path":98,"resourceType":85,"resourceTypeDesc":86,"sizeBytes":99,"sourceHtmlUrl":100},"references\u002Ftroubleshooting.md",9037,"https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fmlops\u002Finference\u002Fvllm\u002Freferences\u002Ftroubleshooting.md",[102,139,164,181,200,223],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":118,"publishedAt":48},"01M71TEDB3XPT9H5R3SH05XWES","writing-plans","Write implementation plans: bite-sized tasks, paths, code.","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Fwriting-plans\u002FSKILL.md",[108,111,114,115],{"categoryNo":109,"categoryName":110,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":112,"categoryName":113,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":23,"categoryName":24,"taxonomyType":25,"taxonomyTypeDesc":26},{"categoryNo":116,"categoryName":117,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000036","内容创作者",[119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138],"writing","writing plans","plans","test-driven","test-driven-development","superpowers","design","workflow","code","tasks","subagent","code-review","review","documentation","write","paths","planning","development","test","implementation",{"skillNo":140,"name":123,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":150,"publishedAt":48},"01M71TEB43BNQKKCRMQZDZ7SAQ","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Ftest-driven-development\u002FSKILL.md",[144,147],{"categoryNo":145,"categoryName":146,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":148,"categoryName":149,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000008","测试与质量",[151,136,137,152,153,154,155,124,119,156,127,129,157,158,159,160,161,121,162,163],"test driven development","driven","green","red-green-refactor","testing","debugging","systematic","refactor","systematic-debugging","enforce","tests","tdd","quality",{"skillNo":165,"name":159,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":167,"categories":168,"keywords":170,"publishedAt":48},"01M71TE76KV9R1WY1V5XQXMMTJ","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Fsystematic-debugging\u002FSKILL.md",[169],{"categoryNo":148,"categoryName":149,"taxonomyType":20,"taxonomyTypeDesc":21},[171,156,157,122,172,173,123,124,119,129,174,175,136,176,121,177,137,178,179,180],"systematic debugging","problem","problem-solving","phase","bugs","root cause","root-cause","investigation","understand","hermes",{"skillNo":182,"name":183,"nameZh":6,"description":184,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":185,"categories":186,"keywords":192,"publishedAt":48},"01M71TE2Y2TRFZNT6YRPKQA33R","subagent-driven-development","Execute plans via delegate_task subagents (2-stage review).","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Fsubagent-driven-development\u002FSKILL.md",[187,188,191],{"categoryNo":112,"categoryName":113,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":189,"categoryName":190,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":116,"categoryName":117,"taxonomyType":25,"taxonomyTypeDesc":26},[193,129,136,152,122,123,194,195,124,119,126,127,130,131,196,197,121,198,199,137],"subagent driven development","delegate","delegation","subagents","parallel","stage","task",{"skillNo":201,"name":202,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":204,"categories":205,"keywords":209,"publishedAt":48},"01M71TE0E7WRGXY6N0WTMDXY23","spike","Throwaway experiments to validate an idea before build.","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Fspike\u002FSKILL.md",[206],{"categoryNo":207,"categoryName":208,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000021","资料检索与研究",[202,210,211,212,135,136,121,213,214,119,215,216,129,217,218,42,219,220,221,222],"concept","exploration","get","throwaway","shit","proof-of-concept","done","prototype","plan","build","sketch","mvp","validate",{"skillNo":224,"name":225,"nameZh":6,"description":226,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":227,"categories":228,"keywords":231,"publishedAt":48},"01M71TDY3JXZHWSYNF3AZ2XM2D","requesting-code-review","Pre-commit review: security scan, quality gates, auto-fix.","https:\u002F\u002Fgithub.com\u002Feliottbusiness\u002FDeptFlow-Agent\u002Fblob\u002Fmain\u002Fprofile\u002Fskills\u002Fsoftware-development\u002Frequesting-code-review\u002FSKILL.md",[229,230],{"categoryNo":145,"categoryName":146,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":23,"categoryName":24,"taxonomyType":25,"taxonomyTypeDesc":26},[232,233,127,131,122,123,234,235,124,236,119,237,238,129,130,239,240,136,241,121],"code review","requesting code review","fix","auto-fix","auto","security scan","commit","verification","security","scan"]