skills.rest
tbvjvsladla
tbvjvsladlaCOMMUNITY
Explore all skills by this author →
GitHub

vllm-recipe-explorer

Generate and optimize LLM recipes for specific GPUs with VRAM constraints.

18|5|Updated Jun 21, 2026
Search Tags:#LLM#recipe generation#tensor parallelism#GPU optimization#VRAM budget
One-click install
npx skills add https://github.com/tbvjvsladla/easy_vllm_simulator --skill vllm-recipe-explorer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vllm-recipe-explorer
Source: https://github.com/tbvjvsladla/easy_vllm_simulator/tree/main/.claude/skills/vllm-recipe-explorer
Command: npx skills add https://github.com/tbvjvsladla/easy_vllm_simulator --skill vllm-recipe-explorer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires vllm, torch, yaml, json, requests, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the generation and optimization of LLM recipes for specific GPUs, streamlining the process of configuring and deploying LLM models.

Core Features & Use Cases

  • Recipe Generation: Automatically generate LLM recipes based on user-defined parameters such as model, GPU, VRAM budget, and quantization.
  • Recipe Optimization: Optimize recipes for target GPUs, considering VRAM utilization, batch size, and attention backend.
  • Use Case: Imagine you have a new LLM model and want to deploy it on an RTX4090 GPU with a 24GB VRAM budget. Use this Skill to generate and optimize a recipe for this specific configuration.

Quick Start

Use the vllm-recipe-explorer skill to generate a recipe for the model 'gpt-oss-20b' on an RTX4090 GPU with a 24GB VRAM budget.

Frequently Asked Questions about vllm-recipe-explorer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I optimize LLM recipes for specific GPUs?▼

You can optimize LLM recipes for specific GPUs by parsing model configurations and simulating LLM performance to automatically calculate tensor parallelism, VRAM utilization, batch size, and attention backend settings.

What is the best way to generate vllm recipes for a 24GB VRAM budget?▼

Generating vllm recipes for a 24GB VRAM budget requires specifying your target GPU like RTX4090 and model parameters, allowing the tool to simulate performance and output an optimized single-node or multi-node deployment configuration.

Does vllm recipe generation support multi-node deployments?▼

Yes, vllm recipe generation supports multi-node deployments by automatically generating and optimizing configurations that account for tensor parallelism, batch size, and VRAM utilization across multiple connected GPUs.

Can I use tensor parallelism optimization for LLMs with torch?▼

Yes, you can use tensor parallelism optimization for LLMs with torch, as the recipe generation process parses model configurations using torch and vllm dependencies to simulate and optimize GPU performance.

Why does my LLM deployment exceed the VRAM budget on my GPU?▼

Your LLM deployment may exceed the VRAM budget if the recipe lacks proper tensor parallelism and batch size optimization, which this tool resolves by simulating performance and automatically adjusting configurations to fit target GPU constraints.

Related Skills

Explore More Software Engineering Skills→
juliuss1907juliuss1907

serving-llms-vllm

Serve LLMs with OpenAI-compatible endpoints and quantization for limited GPU memory.

Community
Advanced
ovachieverovachiever
52

tensorrt-llm

Optimize LLM inference on NVIDIA GPUs with TensorRT-LLM.

Community
Advanced
kngender5kngender5

colab-advanced-inference-optimization

Optimize inference speed and VRAM efficiency for Colab models using FlashAttention-2/3.

Community
Advanced
nelohenriqnelohenriq
1

serving-llms-vllm

Serve large language models with vLLM via OpenAI-compatible endpoints.

Community
Advanced