HackerTrans
TopNewTrendsCommentsPastAskShowJobs

george_123

no profile record

Submissions

Loading Llama-2 70B 20x faster with Anyscale Endpoints

anyscale.com
3 points·by george_123·3 ปีที่แล้ว·0 comments

How continuous batching improves LLM inference throughput 23x

twitter.com
1 points·by george_123·3 ปีที่แล้ว·0 comments

Ant Group – scaling to 1.37M QPS on Ray

anyscale.com
3 points·by george_123·4 ปีที่แล้ว·1 comments

[untitled]

1 points·by george_123·4 ปีที่แล้ว·0 comments

comments

george_123
·3 ปีที่แล้ว·discuss
1) when the blog was released, TGI didn’t support paged attention, 2) many people don’t even know about TGI to reduce inference costs.
george_123
·3 ปีที่แล้ว·discuss
A lot of AWS services (especially SageMaker) aren’t very good in customer experience. People buy them for nominal capabilities and AWS core bread and butter — short-term and long-term reliability.

Most of these startups (AI and others) have to offer a compelling product before even being notable.

Besides, AWS top level doesn’t care if you use sagemaker or not. There’s a premium but if you’re still using EC2 via another startup, they’re still capturing lions share of value.
george_123
·3 ปีที่แล้ว·discuss
this approach to managing KV cache can work with 4bit. imagine the speedup of pagedattention with quantization..
george_123
·4 ปีที่แล้ว·discuss
This is a guest engineering blog post from Ray contributors at Ant Group, discussing how Ant Group implemented scalable Ray Serving architecture atop Ray, deploying 240,000 cores for model serving, scaling by 3.5x from previous year, and reaching 1.37 million TPS during peak times.

(resubmission, URL redirected to /Engineering last time)
george_123
·4 ปีที่แล้ว·discuss
This is a guest engineering blog post from Ray contributors at Ant Group, discussing how Ant Group implemented scalable Ray Serving architecture atop Ray, deploying 240,000 cores for model serving, scaling by 3.5x from previous year, and reaching 1.37 million TPS during peak times.