Skip to content

Directory · Local LLM

vLLM

High-throughput LLM inference engine — PagedAttention, OpenAI-compatible server, multi-GPU serving for production self-host.

Open SourcevLLM (open source) · 🇺🇸 United States

Best for

Production local inference, high QPS on open models, GPU cluster serving

Pros and cons

Pros

  • Best inference throughput for self-hosted models
  • OpenAI-compatible API drop-in
  • Multi-GPU and quantization support

Cons

  • Requires NVIDIA GPUs and ML ops skill
  • Overkill for laptop experimentation
  • Not a chat UI — pair with Open WebUI

Alternatives

OllamaTGI (Hugging Face)LocalAI

More local llm tools

GPT4All

Local LLM · Nomic AI · United States

Open Source

Nomic's open-source desktop app for running local LLMs on consumer hardware, with local file chat via LocalDocs.

Best for

Chatting with local models and documents on everyday laptops

Jan

Local LLM · Menlo Research

Open Source

Open-source ChatGPT-style desktop app that runs models locally via llama.cpp, with an offline-first design.

Best for

A private, open-source local chat app with optional cloud model connections

llama.cpp

Local LLM · ggml.ai

Open Source

C/C++ inference engine for running quantized LLMs in GGUF format efficiently on CPUs and GPUs across platforms.

Best for

Maximum-efficiency local inference and embedding into other software

Browse the full directory →