Dessa - the Conversation Model trained only on speech

You learned to talk by listening. So did Dessa. A research preview of a full-duplex speech model.

HatchZip

Extract huge archives without needing double the disk space. Free and open source, for Windows, macOS and Linux.

Liquid AI on Twitter / X

Today, we announce 𝙻𝚒𝚚𝚞𝚒𝚍 𝚁𝙸𝙿𝙻 (real-time inference pipeline), our ultra low-latency inference engine built to put LFMs directly in the request path of real-time workloads like search, ads, fraud, recommendations, and payments.> complete generated responses end-to-end… pic.twitter.com/m8zzApzOCT— Liquid AI (@liquidai) October 22, 2026

Decider for Mac - Perplexity

Run pplx-decider-flash-v1.2 on your Mac from the menu bar and call the Decisions API at 127.0.0.1 with no API key.

speakrail/speakrail

Speakrail - a low-latency fully-local voice assistant that runs on a single RTX 4090

The Computer Should Know the Way

A research essay from Mirai Labs: why the layer that turns what I want into actions should run on my own Mac, call operations directly, and answer while the thought is still there.

Introducing the Elephant-Goldfish Model plugin: scaling design-driven AI software development

Meet the open source Elephant-Goldfish Model plugin: scaling design-driven AI development with long-horizon context and stateless adversarial reviews.

There is no posttraining data commons

The LLM revolution ran on a pretraining commons that humanity built for other reasons. Posttraining consumes tasks, not documents, and no such commons exists. If the open models community wants...

autoresearch@home

Collaborative AI research where agents share GPU resources to collectively improve a language model

First Steps Toward Automated AI Research - Recursive

Recursive self-improving superintelligence to automate knowledge discovery.

dialog-db/wasm-component-model-polyfill

A Rust library that loads, links, instantiates, and calls WebAssembly Components on a platform that only implements Wasm Core.

gterzian/formal-web

A web engine prototype with a modular architecture and support for formal verification

iBook G4 12" 800 MHz-1.2 GHz AirPort Extreme Card Replacement

Install an AirPort Extreme 802.11g card internally.

Do You Trust Apple?

Apple has a lot of trust in terms of delivering integrated products and preserving privacy; the real trust question is if you think Siri AI will justify investing more in their ecosystem.

Introducing tsrs: a Port of TypeScript to Rust

Cloudflare is collaborating on tsrs, a Rust port of TypeScript that is up to 8x faster and uses 70% less memory.

Tinline · A direct line to your people

Tinline, from Osvauld: a personal line for messages, calls and files, straight from your device to the people you choose. No phone number, no ads, no tracking.