HN
Hacker News
Top
New
Best
Ask
Show
Jobs
george_123
9 karma · joined December 2, 2019
submissions
comments
Loading Llama-2 70B 20x faster with Anyscale Endpoints
anyscale.com
·
3 pts
·
george_123
·
0
How continuous batching improves LLM inference throughput 23x
twitter.com
·
1 pts
·
george_123
·
0
Ant Group – scaling to 1.37M QPS on Ray
anyscale.com
·
3 pts
·
george_123
·
1