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)
Anthropic is accused of treating Russian-speaking Claude users unfairly through a system that charges them before requiring identity verification, leaving some unable to recover subscription fees - especially if they paid with virtual cards or gift subscriptions. The verification process reportedly collects biometric information that may be retained for up to three years, even when access is subsequently denied. The complaints also describe persistent account restrictions: device identifiers and email addresses are allegedly blacklisted, while GitHub repositories connected to an account may become inaccessible. Developers can consequently lose not just Claude but also projects and the ability to transfer or continue their work. Russian emigrants living in Europe are said to face similar blocks when logging in through a Russian IP address.
The central argument is that these measures go beyond geographic or legal restrictions and amount to discrimination based on nationality. The criticism contrasts Anthropic’s approach with Google and OpenAI, which are described as warning users before payment rather than subjecting them to extensive checks afterward. It argues that a U.S. citizen in Russia can use Claude while a Russian citizen legally residing in the EU may be excluded, suggesting an internal company policy rather than a necessary consequence of sanctions. The complaints also point to continued trade in Russian resources through intermediaries as evidence of inconsistent enforcement. Ordinary users, the argument concludes, bear the costs of political decisions through lost money, data, and work tools, while Anthropic has not publicly answered direct questions.
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.