Hackernews
new
show
ask
jobs
Co-Designing AI Models Using Speculative Decoding for Faster LLM Inference
1 points
posted 10 hours ago
by buildbot
(developer.nvidia.com)
No comments yet