Videos
5 total
| Thumb | Title | Channel | Status | Published |
|---|---|---|---|---|
|
Claude Just Shrank Qwen 6x. Then Something Weird Happened
Claude Sonnet 5.5, working alone for 8 hours, produced a compressed 27B Qwen 3.8 model that beats the expert-made... |
Cloud Codes | summarized | 2026-11-09 15:58 |
|
NVIDIA's Secret 2 Bit Model
NVIDIA's Neotron 3.5 Bolt is a 74-billion-parameter mixture-of-experts model with experts quantized to 2 bits,... |
Sam Witteveen | summarized | 2026-11-09 14:00 |
|
Claude Shrank Qwen 27B to 8GB... But It Runs 3x Slower
Claude Sonnet 3.5 produced a 2.5-bit quantized Qwen 27B model that fits in 8GB and beats a human-made quant on... |
Cloud Codes | summarized | 2026-11-09 12:30 |
|
They Removed Quarter of an AI’s Brain. Here’s What Happened......
Pruning six of 36 layers from Qwen 3 with no retraining (LeGO method) leaves short reasoning benchmarks at 94% of... |
Cloud Codes | summarized | 2026-11-05 17:01 |
|
AI on iPhone Offline #shorts
Open BMB claims its 35B parameter Mini CPM V 4.7 runs on an iPhone in 3 GB of memory, but the actual iOS app... |
Cloud Codes | summarized | 2026-10-17 09:42 |
Frontier News · by Hyperjump Technology