Rising1 sources· last seen 6h ago· first seen 6h ago
Show HN: Llama.cpp fork with 2-4x multiGPU speed for MoE models bigger than VRAM
Lead: Hacker NewsBigness: 38metacppfork2-4xmultigpu
📡 Coverage
10
1 news source
🟠 Hacker News
18
1 pts, 2 comments
🔴 Reddit
0
📈 Google Trends
83
Meta AI: 83/100
Full methodology: How scoring works
Receipts (all sources)
Show HN: Llama.cpp fork with 2-4x multiGPU speed for MoE models bigger than VRAM
HACKERNEWS · Hacker News · 6h ago · ▲ 1 · 💬 2
score 144