# abi/secret-llama

Fully private LLM chatbot that runs entirely with a browser with no server needed. Supports Mistral and LLama 3.

Repository: https://github.com/abi/secret-llama
Canonical: https://ross.abutalabs.com/products/secret-llama
Language: TypeScript
License: Apache-2.0
License Family: permissive
Last push: 2024-06-05T02:04:17+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 61
- inputs: {"age_days": 855, "days_push": 820, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2678, forks 169 (observed 2026-08-28T04:07:09.881603+00:00)

## What it is
Secret Llama is a fully private LLM chatbot that runs entirely in the browser with no server or installation required, powered by the WebLLM inference engine. It supports open source models like Llama 3, Mistral, and TinyLlama via WebGPU, works offline, and never sends conversation data off the user's machine.

## Use cases
- chat with an llm without any data leaving my computer
- run llama 3 or mistral locally in the browser
- use a chatgpt-like interface offline with no install
- private ai chatbot with no server
- test open source llm models in chrome with webgpu
- self-host-free local llm chat for sensitive conversations

## When to choose
- privacy is paramount and no data should leave the device
- you want zero installation and no server setup
- you need an offline-capable chatbot in a modern browser
- you want a simple ChatGPT-style UI for open source models

## When to avoid
- you need the largest or newest models beyond browser RAM limits
- your browser lacks WebGPU support
- you need multi-user or server-side features
- maximum inference speed on GPU hardware is required

## Facets
- artifact type: application
- maturity: active
- function: llm-inference, chatbot, chat-interface, wasm
- domain: large-language-models, artificial-intelligence, chatbots, privacy, web-development
- platform: browser, cross-platform
- tags: webgpu, in-browser-llm, no-server, offline, react, webllm, local-inference, web-server

## Member repositories
- abi/secret-llama (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:09.881603+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:17:58.315842+00:00, confidence not recorded.
  - readme: https://github.com/abi/secret-llama (fetched 2026-08-28T04:07:09.881603+00:00, sha 44585dd841da)
- Data as of 2026-08-30T08:39:29.467469+00:00.