Neural Magic Blog company blog
What happens if we remove 50 percent of Llama?
Discover Sparse Llama: A 50% pruned, GPU-optimized Llama 3.1 model with 2:4 sparsity, enabling faster, cost-effective inference without sacrificing accuracy.
- Points
- 231
- HN comments
- 132
- Submissions
- 1
- Cited by
- 0 commenters
AI & Machine LearningDatabases & Data2024
On Hacker News
| Submitted | Title on HN | Points | Comments |
|---|---|---|---|
| Nov 26, 2024 | What happens if we remove 50 percent of Llama? by BUFU | 231 | 132 |
