llmspy.org

llmspy.org is an open source, self-hosted web UI, CLI, and server gateway that unifies access to local and cloud LLMs.

llms.py is an open source, multi-user ChatGPT and OpenRouter alternative that functions as a self-hosted gateway, web UI, and CLI. It lets users chat, generate images, synthesize audio, and execute code using over 530 local and cloud models across 24 providers. The application supports text-to-speech, voice-to-text input, and image generation using providers such as Gemini, OpenAI, OpenRouter, Chutes, Z.ai, and Nvidia. It also includes support for Model Context Protocol (MCP) servers, Python function calling, RAG document stores, and isolated workspaces for AI coding agents. Users can deploy citation-backed assistants to websites via a script tag or publish creations to a public gallery.

llmspy.org at a glance

Best for: Software engineer
Runs on: Web, CLI, self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI chat assistant
Website: llmspy.org

Key features

Access 530+ models from 24 cloud and local providers
Generate images via Gemini, OpenAI, OpenRouter, Chutes, Z.ai, and Nvidia
Synthesize speech with TTS and capture input via voice-to-text transcription
Connect to Model Context Protocol (MCP) servers and Python function calls
Execute Python, JavaScript, TypeScript, and C# code directly in the UI
Run RAG workflows by ingesting files, folders, and crawled websites
Design PDFs using Typst templates with schema-driven forms and AI assistance
Configure specialized AI agents with isolated filesystem access to project folders

Use it for

Chatting with local and cloud LLMs via a web interface or command line
Generating images and audio using multiple AI model providers
Running AI coding and planner agents within isolated project folders
Deploying citation-backed website support assistants with a script tag
Creating and previewing custom PDF documents using Typst templates

Companions

Tools that articles about llmspy.org recommend using with it, each with its job.

Cline — Open source, any model, with approval on every step
CrewAI — Point CrewAI's agents at the local endpoint.
SearXNG — Built-in integration for web-grounded answers.
ComfyUI — Local image and video generation on a node canvas, with every model and parameter in reach
Hermes Agent — Adds agentic workflows on the same local models.
Cloudflare Tunnel — Reach the homelab without opening ports.

Alternatives to llmspy.org

LLM CLI — It enables command-line interaction with remote APIs and local machine models.
Chat UI — It runs instantly anywhere with a web browser and weighs under 100 KB.
LM Studio — Runs open source language models locally with real-time voice transcription.
LLMStack — Builds AI applications and agents with custom data without writing code.
Open WebUI — Run AI on your own terms with complete data privacy and self-hosting control.
LiteLLM — One OpenAI-format API for 100+ LLM providers
MLX LM — Runs LLMs and fine-tuning natively on Apple silicon hardware.
Ollama — Run open models locally or in the cloud with private prompts
StrictLLM — It keeps your AI workflows secure and reviewable with sandboxed MCP servers and strict tool permissions.
LibreChat — Unifies multiple AI models and tools in one customizable, open-source platform.
llmware — Build private, local RAG pipelines with 300+ prepackaged models and integrated document parsing.
Text Generation WebUI — Run local LLMs with multiple backends, multimodal input, and tool calling 100% offline.

All llmspy.org alternatives