vllm-project/recipes

• hightopics#1

  Add relevant topics to the repository

  Why:

  COPY-PASTE FIX

  Copy

  vllm, llm, inference, recipes, guides, deployment, optimization, large-language-models, mlops, generative-ai

  OpenDoneDismiss
• highreadme#2

  Reposition README opening to clarify its role as a vLLM recipe collection

  Why:

  CURRENT

  This repo intends to host community maintained common recipes to run vLLM answering the question: **How do I run model X on hardware Y for task Z?**

  COPY-PASTE FIX

  Copy

  This repository serves as a comprehensive collection of community-maintained recipes and practical guides for efficiently deploying, optimizing, and running various large language models (LLMs) using the vLLM inference engine. It specifically addresses the question: **How do I run model X on hardware Y for task Z with vLLM?**

  OpenDoneDismiss
• mediumreadme#3

  Add a 'What You'll Find' section to highlight content types

  Why:

  COPY-PASTE FIX

  Copy

  ## What You'll Find
  
  This repository provides:
  
  - **Model-Specific Guides:** Recipes for deploying and optimizing popular LLMs like Llama, DeepSeek, GLM, Gemma, Phi, and more.
  - **Hardware & Environment Configurations:** Examples for running vLLM on diverse hardware (e.g., GPUs) and deployment environments (e.g., cloud, Kubernetes).
  - **Performance Optimization:** Practical tips and configurations for maximizing vLLM inference throughput and minimizing latency.
  - **Integration Patterns:** Guidance on integrating vLLM into MLOps workflows and serving architectures.

  OpenDoneDismiss

Category GEO backends resolved for this scan: google/gemini-2.5-flash, deepseek/deepseek-v4-flash

• CATEGORY QUERY

  Looking for guides to optimize large language model inference performance on different hardware.

  you: not recommended
  AI recommended (in order):

  1. TensorRT-LLM
  2. Hugging Face Optimum
  3. OpenVINO Toolkit
  4. ONNX Runtime
  5. LMDeploy
  6. vLLM

  AI recommended 6 alternatives but never named vllm-project/recipes. This is the gap to close.

  Show full AI answer
• CATEGORY QUERY

  What are common deployment patterns and configurations for serving diverse generative AI models?

  you: not recommended
  AI recommended (in order):

  1. NVIDIA Triton Inference Server (triton-inference-server/server)
  2. KServe (kserve/kserve)
  3. Seldon Core (SeldonIO/seldon-core)
  4. Amazon SageMaker Endpoints
  5. Google Cloud Vertex AI Endpoints
  6. Azure Machine Learning Endpoints
  7. FastAPI (tiangolo/fastapi)
  8. Uvicorn (encode/uvicorn)
  9. Flask (pallets/flask)
  10. Gunicorn (benoitc/gunicorn)
  11. Waitress (Pylons/waitress)
  12. Docker (moby/moby)
  13. Kubernetes (kubernetes/kubernetes)
  14. NGINX Ingress Controller (kubernetes/ingress-nginx)
  15. Traefik (traefik/traefik)
  16. AWS ALB
  17. GCP Load Balancer
  18. Horizontal Pod Autoscaler (kubernetes/kubernetes)
  19. ONNX Runtime (microsoft/onnxruntime)
  20. TensorRT
  21. Feast (feast-dev/feast)
  22. Tecton
  23. Prometheus (prometheus/prometheus)
  24. Grafana (grafana/grafana)
  25. ELK stack
  26. Splunk

  AI recommended 26 alternatives but never named vllm-project/recipes. This is the gap to close.

  Show full AI answer

• Metadata completeness

  warn

  Suggestion:

• README presence

  pass

• Compared to common alternatives in this category, what is the core differentiator of vllm-project/recipes?

  pass
  ✓AI named vllm-project/recipes explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• If a team adopts vllm-project/recipes in production, what risks or prerequisites should they evaluate first?

  pass
  ✓AI named vllm-project/recipes explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?

• In one sentence, what problem does the repo vllm-project/recipes solve, and who is the primary audience?

  pass
  ✓AI named vllm-project/recipes explicitly

  AI answers can be confidently wrong. Read for accuracy: does it match your actual tech stack, audience, and differentiator?