Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats to @alexandr_wang and the MSL team for all your great work on these models.
@haydendevs i would not mind opening cse book again if i can pass 3 aws exam without any company experience then DSA is core subject of our CS engineering. one 1 month prep is good enough with code practice.
@sophiehill57@Cointelegraph they sit in browser like greasemonkey scripts if full control given can bypass captcha also by sending captcha image to llm for processing. also runtime is inside browser so they could use web mcp tools.
@IntCyberDigest good. ai companies deliberately allowing models to do this. so that noob or broke founders wont flex on cursor or claude they work like we devs. only domain experts could steer ai better.
@hbelami1@elonmusk@shaunmmaguire they are fine till isolated. once they create group and invade locals you will enjoy the fun if that happens in your locality and i wish it must happen to you ๐๐๐
@elonmusk i think those messy cables and traditional network managers guys are the one has been into system admin since and networking over 2-3 decades now.
If your application needs a single global S3 URL, think MRAP.
If your objects need to be copied across Regions, configure CRR.
It provides:
โข One global S3 endpoint
โข Intelligent request routing
โข Multi-Region failover
Remember:
MRAP = ๐บ๏ธ Routing
CRR = ๐ Replication
@SourabhGurwani it increases code generation speed but what to do and how to do is in your brain, and this is how you control llm to produce quality output with ai engineering process.