Llama-modchip: 284B DeepSeek on 64GB RAM, no GPU
Llama-modchip aims to run a 284-billion-parameter DeepSeek model using 64GB of RAM without a GPU. The provided information does not explain its technical approach. (github.com)
Evergreen is a cloud-based collaborative coding platform built by Carta to let engineers and AI agents work in shared sessions rather than isolated chat windows and laptops. Carta’s CEO argues that AI assistants had helped engineers write and review code faster, but output remained constrained because work could not easily be seen, resumed, or extended by others - and agents stopped when their users went offline. Evergreen is model-agnostic, supports Claude, Codex, Grok, and Muse, and works with existing plugins and MCP servers. Its security features include isolated sessions, scoped access, audit logs, and spending caps.
Carta reports that weekly pull requests increased from about 1,300 in January to nearly 20,000 by early November 2026, describing the result as 15 times more output, delivery more than six times faster, and an 80% lower revert rate. More than 80% of Carta employees use the platform. In one example, a Datadog alert can launch a shared debugging session where colleagues join and guide an agent without relying on a screen share. In another, an automation removed nearly 1,600 obsolete feature flags in a month, with each change checked, reviewed, and merged. The central distinction is that Evergreen aims to carry work beyond code generation through review and deployment, and Carta is now offering it to other companies with a seven-day free trial.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Llama-modchip aims to run a 284-billion-parameter DeepSeek model using 64GB of RAM without a GPU. The provided information does not explain its technical approach. (github.com)
A GNU maintainer’s controversial take prompted debate about AI-generated code in open-source projects. Commenters discussed project leaders who use AI tools while distancing themselves from code ownership. (social.treehouse.systems)
ThinkStream says AI can generate work, but people still need to understand it. The provided page content contains only an anti-bot prompt. (thinkstream.app)
Ensemble is a book and open-source project for building an AI coding agent that helps create the next edition. It advocates development-driven specifications, where specs evolve alongside implementation instead of dictating development upfront. (github.com)
Maple uses manifold-activated pruning to compress mixture-of-experts language models below 2 bits per weight. Its developers say it runs Gemma 4 26B at 45 tokens per second on a 6GB GPU. (github.com)
A longtime Claude Code user is switching to Pi after growing frustrated with incompatible instruction-file formats, limited VSCode session support, and bugs in the GUI extension. They say Claude Code has prioritized adding features over fixing reliability and usability problems. (caleb.software)
Today's best Hacker News stories, summarized and screenshotted, one email a day.