Llama | search

Overview
Newspapers
Aggregators
Blogs
Videos
Photos
Websites

刚试了一下，用llama.cpp运行4bit量化的Qwen-72B-Chat，生成速度是5 tokens/s左右。另外，llama-70b不是一个性价比高的选择，mistral 7b以及国产的qwen 14b/baichuan 13b效果也都挺不错的，稍微SFT一下（从GPT4搞几百条数据蒸馏就够了）， More @Wikipedia

Get the latest news about Llama from the top news sites, aggregators and blogs. Also included are videos, photos, and websites related to Llama.

Hover over any link to get a description of the article. Please note that search keywords are sometimes hidden within the full article and don't appear in the description or title.

Llama | search

Llama Info

Llama Photos

Llama Websites

Llama Videos

CNN »

NEW YORK TIMES »

FOX NEWS »

THE ASSOCIATED PRESS »

WASHINGTON POST »

AGGREGATORS

GOOGLE NEWS »

YAHOO NEWS »

BING NEWS »

ASK NEWS »

HUFFINGTON POST »

TOPIX »

BBC NEWS »

MSNBC »

REUTERS »

WALL STREET JOURNAL »

LOS ANGELES TIMES »

BLOGS

FRIENDFEED »

WORDPRESS »

GOOGLE BLOG SEARCH »

YAHOO BLOG SEARCH »

TWINGLY BLOG SEARCH »