Skip to main content

ZeroChat

English · Español


English

ZeroChat is a personal, private web client for chatting with AI models, consulting documents and running tools. This documentation covers the 10.x series. Free and open source under the MIT licence.

The interface runs entirely in the browser as static files: history and settings remain in local browser storage without automatic cloud synchronization.

Key features

  • AI Profiles: OpenAI and compatible chat APIs, Ollama, OpenRouter, Claude and Gemini. Supports reasoning, vision and function calling.
  • WebLLM: In-browser WebGPU inference with downloadable models (requires compatible hardware).
  • Documents & RAG: Local document attachments and collections with client-side text extraction and Orama lexical search (no external embeddings required).
  • Agentic Tools: JavaScript sandbox, web search, page fetching, interactive charts and read-only subagents. The optional local backend adds filesystem access, shell commands, diagnostics and MCP services.
  • Diagnostics: Conversation branches, JSON and Markdown export, PDF print, context metrics and request inspector.

Quick start

  1. Web (client-only): Open ZeroChat on GitHub Pages, configure a profile, and start chatting.
  2. With local computer tools (Python backend):
# Via PyPI (Linux, macOS, Windows)
pip install zerochat
zerochat

# Or download and run directly
curl -fsSL https://albalday.github.io/zerochat/zerochat.py -o zerochat.py
python3 zerochat.py

Documentation & Help

Comprehensive guides covering configuration, supported environments (Linux, Windows, macOS, Android/Termux), MCP services, and security:

Development

ZeroChat is developed on the dev branch:

git checkout dev
npm install
npm test

For development rules, architecture and test suites, see AGENTS.md, js/README.md, py/README.md and tests/README.md.


Español

ZeroChat es un cliente web de uso personal para conversar con modelos de IA, consultar documentos y ejecutar herramientas. Esta documentación corresponde a la serie 10.x. Proyecto de libre uso bajo licencia MIT.

La interfaz se ejecuta íntegramente en el navegador como archivos estáticos: el historial y la configuración se conservan en el almacenamiento local del navegador, sin sincronización automática en la nube.

Características principales

  • Perfiles de IA: OpenAI y APIs compatibles, Ollama, OpenRouter, Claude y Gemini. Soporta razonamiento, visión y llamadas a herramientas (tool calling).
  • WebLLM: Inferencia local en el navegador mediante WebGPU tras descargar el modelo (requiere hardware compatible).
  • Documentos y RAG: Adjuntos y colecciones locales con extracción de texto en el cliente y búsqueda léxica mediante Orama (sin requerir embeddings externos).
  • Herramientas agénticas: Entorno aislado de JavaScript, búsqueda y lectura web, gráficos interactivos y subagentes de solo lectura. El backend local opcional añade acceso al sistema de archivos, ejecución de comandos, diagnósticos y servicios MCP.
  • Diagnóstico e historial: Ramas de conversación, exportación a JSON y Markdown, impresión en PDF, métricas de contexto e inspección de peticiones.

Inicio rápido

  1. Web directa (solo cliente): Abre ZeroChat en GitHub Pages, configura un perfil y comienza.
  2. Con herramientas del equipo (backend Python):
# Mediante PyPI (Linux, macOS, Windows)
pip install zerochat
zerochat

# O descarga directa del script
curl -fsSL https://albalday.github.io/zerochat/zerochat.py -o zerochat.py
python3 zerochat.py

Documentación y ayuda

Guías completas de configuración, soporte de entornos (Linux, Windows, macOS, Android/Termux), servicios MCP y seguridad:

Desarrollo

El desarrollo se realiza en la rama dev:

git checkout dev
npm install
npm test

Las normas de desarrollo, arquitectura y suite de pruebas se describen en AGENTS.md, js/README.md, py/README.md y tests/README.md.

Metadata

Release files for zerochat 10.20.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for zerochat 10.20.0
File Size Uploaded Provenance Actions
zerochat-10.20.0.tar.gz 127.8 kB With provenance Details for zerochat-10.20.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for zerochat 10.20.0
File Interpreter ABI Platform Actions
zerochat-10.20.0-py3-none-any.whl Python 3 none any Details for zerochat-10.20.0-py3-none-any.whl

Total release size: 256.5 kB

Release files / zerochat-10.20.0.tar.gz

Download URL zerochat-10.20.0.tar.gz
Size 127.8 kB
Tags Source
SHA-256 checksum
How to use checksums
6f9db311727a129838aeff3daa53eab0d9e9388406bafda4e5f07f02c0d0d48b
BLAKE2b-256 checksum
How to use checksums
d419f2cba0e9ba1243b0db41a3e11e3fe4fdd6baa441eb508011b960fc5614b0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 15, 2026.

Transparency log

Release files / zerochat-10.20.0-py3-none-any.whl

Download URL zerochat-10.20.0-py3-none-any.whl
Size 128.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
d203f5a9ebf92b43c6345e62262dcc696d6dc4bc3f44fe1f57ce1d6f9e5ab15d
BLAKE2b-256 checksum
How to use checksums
4578db5b0f62dcd766ab4f1d670e30ebfba48b97f3a455a59dcde7078b3622d4
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 15, 2026.

Transparency log

Security for zerochat 10.20.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for zerochat-10.20.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 15, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

10.20.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page