LLMs & Chat Models
Large language models, chat assistants, and the runtimes that serve them. From flagship hosted models like Claude, GPT, and Gemini to local-first runners like Ollama, LM Studio, vLLM, and Llamafile. As the model landscape evolves we'll keep growing this category.
LM Studio
Desktop app to run AI models locally and privately. Headless server mode, JS/Python SDKs, supports gpt-oss, Qwen, Gemma, DeepSeek.
Jan
Open-source ChatGPT-style desktop assistant with 4M+ downloads. Personal intelligence that answers only to you.
vLLM
High-throughput, memory-efficient open-source LLM inference and serving engine. Production-grade, originally from UC Berkeley.
LocalAI
OpenAI/Anthropic-compatible REST API for running LLMs, agents, and document intelligence locally on your hardware.
Text Generation WebUI
The original Gradio web UI for running local LLMs. 100% private, 100% offline, multi-backend, multimodal.
Llamafile
Mozilla.ai project that distributes and runs LLMs as single-file executables across any OS using Cosmopolitan Libc.
6 tools in this category