Optional package providing compatibility with older browsers for @wllama/wllama

WebAssembly binding for llama.cpp - Enabling on-browser LLM inference

wllama provider for @localmode - GGUF model inference via llama.cpp WASM with metadata inspection and browser compatibility checking

Optional package providing compatibility with older browsers for @wllama/wllama

Drive the LFM2.5 (230M/350M) frontend-agent model in the browser: tool-calling + RAG-grounded answering, fully typed. Loads the GGUF from Hugging Face via wllama.

Framework-agnostic browser LLM service wrapper with WebGPU and WebAssembly support

On-device LLM sessions for any browser — Chrome's Prompt API, WebLLM, or wllama, picked by feature detection, smallest model first.

WebAssembly binding for llama.cpp - Enabling on-browser LLM inference

WebAssembly binding for llama.cpp - Enabling on-browser LLM inference

WebAssembly binding for llama.cpp (with Q2_0 ternary support) - Enabling on-browser LLM inference

Custom WebAssembly binding for llama.cpp - Enabling on-browser LLM inference

WebAssembly binding for llama.cpp - Enabling on-browser LLM inference (custom version from aviallon)

A browser-friendly library for running LLM inference using Wllama with preset and dynamic model loading, caching, and download capabilities.

Local-first AI utilities for the browser. Zero dependencies. Privacy-first.

WebAssembly package for llama.node browser inference

## Download and run local LLMs within your browser.

Run language models in the browser

Chrome AI provider for @localmode - on-device inference via Chrome's built-in Gemini Nano

A reflex layer for every input on the web. Drop in SmartField / SmartForm / SmartText and get on-device autofill, contextual spellcheck, smart paste, and cross-field inference — zero latency, zero cost, zero privacy exposure.

The decision layer for in-browser AI: given a model id, tells you if this browser can run it, over which backend, which quant variant, and exactly how many bytes it will download.