Launch
Muse Ember: Our smallest, most efficient model (opens research.meta.ai)
Meta's entry-level model, described by SiliconANGLE as costing significantly less than the competition and the first trained using Muse Wonder.
DiscussionUnverified claim
Llama-modchip: 284B DeepSeek on 64GB RAM, no GPU
- Source
- GitHub (github.com)
- Published
- First seen here
- Edition
- Sunday, November 8, 2026, ranked 10
- Status
- A claim made in a post or discussion that nobody independent has verified.
- Engagement
- As shown by each platform when we last looked ().
- Why it’s here
- 475 points on Hacker News for running a 284B model without a GPU; the performance claim is the author's.
- Topics
- AI, Open source
