pandada8

5 ms·

1.
by pandada8 11mo ago
very curious, which model can only run up to 950 tok/s even with cerebras.
2.
by pandada8 11mo ago
Do most CUA agents use image input? How much latency difference is there between image and DOM representations?
3.
Actual LLM agents are coming. They will be trained(vintagedata.org)
2 points by pandada8 2y ago|0 comments
4.
New tools for building agents with the API [video](youtube.com)
3 points by pandada8 2y ago|0 comments
5.
by pandada8 2y ago
https://fireworks.ai/blog/cursor Fireworks AI has a blog about it.
6.
by pandada8 4y ago
there seems some random 503 in registry.yarnpkg.com too
7.
by pandada8 5y ago
https://github.com/wez/wezterm
8.
by pandada8 7y ago
Dell's DA300 can do 4k 60Hz with DP1.4 though. And intel added dp1.4 support starting with Ice Lake.
9.
UKSM: Swift Memory Deduplication [pdf](usenix.org)
4 points by pandada8 8y ago|1 comments
10.
by pandada8 9y ago
20:55 CST Everything operating normally.
11.
by pandada8 9y ago
Now I am curious about the reason and how github put a fix with this outage
12.
by pandada8 9y ago
seems the error rate is increasing I guess the status page is actually updated manually while the data is auto updated
13.
GitHub.com down?(github.com)
56 points by pandada8 9y ago|34 comments
14.
by pandada8 10y ago
same here :(
15.
by pandada8 10y ago
https://status.github.com/messages/2016-07-21 by the way, why the status from July 15, 2016 to July 21, 2016 is red ?
16.
by pandada8 10y ago
any source?
17.
by pandada8 11y ago
What's difference between this one and rsync？
18.
by pandada8 11y ago
actually in china many big sites are still using http in most occasion