@csb @adam "the company believes that about 7% of users will now hit session limits they wouldn’t have before"
-
@csb @adam "the company believes that about 7% of users will now hit session limits they wouldn’t have before"
Based on what I see some people describing as their use case it seems they are probably trying to curb the extreme power users that open multiple agents at once in multi-hour Ralph loops. That 7% tracks as the amount of *abusive* users a service like that would probably encounter.
That being said, I expect we will see price increases in the future. Hefty ones. Drug dealer model.
-
@csb @adam "the company believes that about 7% of users will now hit session limits they wouldn’t have before"
Based on what I see some people describing as their use case it seems they are probably trying to curb the extreme power users that open multiple agents at once in multi-hour Ralph loops. That 7% tracks as the amount of *abusive* users a service like that would probably encounter.
That being said, I expect we will see price increases in the future. Hefty ones. Drug dealer model.
Dave, I disagree, that price increases of cloud LLM usage will happen: because of competition from Chyna and because of techniques like TurboQuant: https://research.google/blog/turboquant-redefining-ai-efficiency-with-extreme-compression/
in meantime we all need to keep an eye on URL that shows current hitherto usage
-
I podcastindex.social shared this topic
-
Dave, I disagree, that price increases of cloud LLM usage will happen: because of competition from Chyna and because of techniques like TurboQuant: https://research.google/blog/turboquant-redefining-ai-efficiency-with-extreme-compression/
in meantime we all need to keep an eye on URL that shows current hitherto usage
-
Dave, I disagree, that price increases of cloud LLM usage will happen: because of competition from Chyna and because of techniques like TurboQuant: https://research.google/blog/turboquant-redefining-ai-efficiency-with-extreme-compression/
in meantime we all need to keep an eye on URL that shows current hitherto usage
@csb @adam Turboquant is just math. Now your local model has 4X the context window, but the cloud models will also. Yes, local models will become more capable but the cloud models will also improve one to one so I don’t think it’s going to change the equation at all on the larger business side of things.
Companies like anthropic are bleeding cash. They will have to raise prices. I don’t see how higher token compression changes that.
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register LoginWelcome To Podcasting.Chat!
This forum is for podcasters, podcast guests, and podcast enthusiasts alike to share tips, tricks, and their love of the medium.
This forum is fully federated, so you are able to contribute to any discussion here through your own software of choice (e.g. Mastodon, Misskey, Lemmy, Piefed, etc.). So you can sign up for an account here and it federates around the Fediverse. You can also follow feeds and topics from your other Fedi-enabled accounts.
Recent Posts
-
Punjabi Pioneers
rings to light the voices of Punjabi women from early Sikh families in Far North Queensland, whose courage, resilience, and contributions helped shape Punjabi life in Australia...Great Australian Pods Podcast Directory: https://www.greataustralianpods.com/punjabi-pioneers/
#AusPods #Podcasts #Podcasting #Australia #Society #Culture #Identity #History #Language #Diversity #Punjabi
-
@dave Would you consider the hosted API itself to be a critical element of what the index is doing for the community? One could make the case that the open-source code for anyone to run it themselves is the better use of your time than maintaining a free service for people who’re never going to appreciate the v4v model.
Imagine if the API had trial-sized rate limits, then flipped into a pay-what-you-want model powered by L402/x402
-
Remembering Them Podcast
A podcast for guests to speak about their losses, experiences with grief, and to share stories of their loved ones...Great Australian Pods Podcast Directory: https://www.greataustralianpods.com/remembering-dad/
#AusPods #Podcasts #Podcasting #Australia #Society #Culture #Families #PersonalStories #Funerals
-
@nathan That dropped it a lot. Thanks brother.
