localStorage

Evidence (10)

• 5Hacker News·7/31/2026open_source
  Run Kimi K3 with 29 GB of RAM at 0.50 tokens per second

  A GitHub project shows a way to run the Kimi K3 model locally using about 29 GB of RAM, albeit at a slow 0.50 tokens per second. The Hacker News discussion focuses on the practicality of running a large model on modest hardware and the tradeoff between memory usage and inference speed.

• 4Hacker News·7/28/2026open_source
  Running Kimi K3 on an M1 Max

  The post shows an attempt to run Kimi K3 on an Apple M1 Max machine, highlighting practical local inference and performance considerations on consumer hardware. The discussion focuses on feasibility, efficiency, and community interest in running large models outside the cloud.

• 6Simon Willison·7/20/2026research
  Altman Email Reveals OpenAI’s Local GPT-3-Scale Model Plan

  An email from Sam Altman to OpenAI’s board in October 2022 said the company wanted to soon release a language model with roughly GPT-3-level capability that could run locally on consumer hardware. The email said the move was partly intended to discourage rivals from releasing similar models and to make it harder for new efforts to attract funding.

• 2Simon Willison·7/17/2026product_launch
  LLM cliché highlighter

  Simon Willison highlighted a browser-based tool that detects and marks common clichéd phrases often found in LLM-generated text, such as "no X, no Y" chains and "you already know." The app shows pattern counts and matches, supports quick navigation, and stores settings locally in the browser.

• 4InfoQ·7/13/2026framework_update
  Panel Explores Challenges of Sovereign Data and Local-First Computing

  A panel on data ownership argued that control over an account is not enough to count as true ownership. Speakers said user sovereignty also depends on structural independence, interoperability, shared standards, and community governance.

• 4InfoQ·7/13/2026framework_update
  AT Protocol for Resilient Local-First Apps

  Jake Lazaroff presented AT Protocol as a foundation for distributed applications beyond social networking. He described a local-first approach where user data stays in personal data servers while shared infrastructure handles synchronization and updates, reducing dependence on app-specific backends.

• 5Hacker News·7/3/2026open_source
  Guide to Running State-of-the-Art LLMs Locally

  This GitHub guide explains practical ways to run state-of-the-art large language models locally. It focuses on setup choices, hardware considerations, and tooling for people who want to use LLMs without relying on hosted APIs.

• 6Hacker News·7/2/2026regulation
  Right to Local Intelligence

  This item highlights a proposal advocating for the ability to run intelligence and compute locally rather than depend entirely on remote cloud services. It frames local execution as a matter of user rights, control, and resilience, especially as AI systems become more central to digital products and services.

• 4InfoQ·6/29/2026research
  Adam Wiggins on Local-First Architecture and Hybrid AI

  In this podcast episode, Heroku co-founder and Ink & Switch founder Adam Wiggins advocates for a local-first architecture that blends cloud collaboration with the speed and data ownership of local software. He also discusses using CRDTs and version control concepts beyond code, along with a hybrid AI approach that may rely on local models for core productivity tasks.

• 8Hugging Face Blog·2/20/2026open_source
  GGML and llama.cpp join Hugging Face

  Hugging Face announced that GGML, the team behind llama.cpp, is joining the company to support the long-term growth of local AI. The project will remain open source and community driven, with the team retaining technical autonomy while Hugging Face provides sustainable resources and packaging improvements.