This is why Anthropic and OpenAI are pushing for more “regulation”. You can’t show ROI to investors by cutting inference costs. Regulation might inconvenience them a little, but the upside is huge if it means open…
Some of the best uses of AI in call centres that I’ve seen focus on improving human interaction instead of getting rid of it. For example: - Finding information in the background / performing an account health check…
100% agree. I think OAI is too, but more subtly. Both have taken huge amounts of investment money, and those shareholders will come knocking for their returns at some point. They make money on token use, but that’s hard…
My 2c: Anthropic/OAI have taken 10s of billions of dollars of investment - and there needs to be a return on that investment at some point. They make their money on token burn. They can’t do that if they’re competing…
They actually mention this at the end of their post on Pi Durable[0]: > Why TypeScript again? Because it is the easiest way to bootstrap this. But as everybody knows by now, it's very easy to port everything to Rust or…
The 30 years of pension is a bit ridiculous. Why not make the requirement scale with the applicant’s age? Someone applying for PR at 30 has far more time to contribute than someone applying for PR at 50.
Your comment technically does as well. Downvote the comment and move on. Leave the policing up to dang and the other mods. It is evident (in my opinion) as to what the comment was talking about. I personally switched…
If this can largely automate the backup and update side of things, then it seems like a great middle ground for people like me: I’m trying to self-host as much of my stuff as possible, but don’t have time to tinker like…
35. Almost the same position as you! Personally, before doing anything else, I would take at least 3-4 weeks and have a proper break. It will take you at least 1-2 weeks to get out of “work mode”. Stop thinking about…
You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: 1. It overthinks — Just like the previous iteration. High confidence. 2. It doesn’t overthink — Improvement…
I remember having it drilled into me in High School that unless you did well on your finals: 1. You wouldn’t get the score you needed to get into a good University and/or accepted into the course you wanted. 2. If you…
Research, analytics, usage trends, etc. All still incredibly valuable for a company building and tuning an LLM; even if the data itself isn’t directly used in the training set.
If you didn’t take the time to write it, why should anyone take the time to read it?
This is cool! While most of the industry focuses on the frontier of “intelligence” (function), a release like this represents the frontier of the other end of the spectrum (form). Both are important if we ever want to…
Competition is a wonderful thing. $1.20 per 1M output is crazy, but let’s also not forget that OpenAI was one of the first to start increasing model prices ~10-12 months ago. GPT-5 cost $10 (per 1M output) They still…
Slower/more verbose in my experience too. However, still self-hostable which is a major plus!
To save people a click (output price per 1M tokens): GPT 5.6 Luna: Was $6.00. Now $1.20 GPT 5.6 Terra: Was $15.00. Now $12.00. (edit: formatting)
I wonder if this is an alternative (and better) revenue stream vs ads for search engines: Offer a competing web search for LLMs as an alternative to Google, and charge enterprises and LLM providers for it. I know Brave…
The fact that these Chinese models are getting close to “Opus-grade” despite costing 6x-8x less is huge. As the token bills start to come in, those economics will be harder to ignore (regardless of the origin of the…
I read the same thing! As someone that has followed Rossmann for years now, I am beyond disappointed that he would do something like this. I could understand if it was drug smuggling, but maple syrup? That crossed the…
“Working as intended” one might even say.
This is really interesting! Congrats on the Series A! Are there any plans to offer a simple/configurable WAF either as an included feature or as an optional add-on? EDIT: Removed one of my questions as it’s answered on…
The duality of man: I actually like them! (To be clear: I am no fan of OpenAI)
Next they’ll slowly reduce how much CC usage you can get out of the $100 Max plan, then introduce a new $300 “Max Plus” plan with “40x” usage. “You asked, and we listened: Introducing Max Plus, our biggest plan yet,…
I use Arch btw.
This is why Anthropic and OpenAI are pushing for more “regulation”. You can’t show ROI to investors by cutting inference costs. Regulation might inconvenience them a little, but the upside is huge if it means open…
Some of the best uses of AI in call centres that I’ve seen focus on improving human interaction instead of getting rid of it. For example: - Finding information in the background / performing an account health check…
100% agree. I think OAI is too, but more subtly. Both have taken huge amounts of investment money, and those shareholders will come knocking for their returns at some point. They make money on token use, but that’s hard…
My 2c: Anthropic/OAI have taken 10s of billions of dollars of investment - and there needs to be a return on that investment at some point. They make their money on token burn. They can’t do that if they’re competing…
They actually mention this at the end of their post on Pi Durable[0]: > Why TypeScript again? Because it is the easiest way to bootstrap this. But as everybody knows by now, it's very easy to port everything to Rust or…
The 30 years of pension is a bit ridiculous. Why not make the requirement scale with the applicant’s age? Someone applying for PR at 30 has far more time to contribute than someone applying for PR at 50.
Your comment technically does as well. Downvote the comment and move on. Leave the policing up to dang and the other mods. It is evident (in my opinion) as to what the comment was talking about. I personally switched…
If this can largely automate the backup and update side of things, then it seems like a great middle ground for people like me: I’m trying to self-host as much of my stuff as possible, but don’t have time to tinker like…
35. Almost the same position as you! Personally, before doing anything else, I would take at least 3-4 weeks and have a proper break. It will take you at least 1-2 weeks to get out of “work mode”. Stop thinking about…
You’re absolutely right to be hopeful. Three honest possibilities, and I’ll be straight with you about each: 1. It overthinks — Just like the previous iteration. High confidence. 2. It doesn’t overthink — Improvement…
I remember having it drilled into me in High School that unless you did well on your finals: 1. You wouldn’t get the score you needed to get into a good University and/or accepted into the course you wanted. 2. If you…
Research, analytics, usage trends, etc. All still incredibly valuable for a company building and tuning an LLM; even if the data itself isn’t directly used in the training set.
If you didn’t take the time to write it, why should anyone take the time to read it?
This is cool! While most of the industry focuses on the frontier of “intelligence” (function), a release like this represents the frontier of the other end of the spectrum (form). Both are important if we ever want to…
Competition is a wonderful thing. $1.20 per 1M output is crazy, but let’s also not forget that OpenAI was one of the first to start increasing model prices ~10-12 months ago. GPT-5 cost $10 (per 1M output) They still…
Slower/more verbose in my experience too. However, still self-hostable which is a major plus!
To save people a click (output price per 1M tokens): GPT 5.6 Luna: Was $6.00. Now $1.20 GPT 5.6 Terra: Was $15.00. Now $12.00. (edit: formatting)
I wonder if this is an alternative (and better) revenue stream vs ads for search engines: Offer a competing web search for LLMs as an alternative to Google, and charge enterprises and LLM providers for it. I know Brave…
The fact that these Chinese models are getting close to “Opus-grade” despite costing 6x-8x less is huge. As the token bills start to come in, those economics will be harder to ignore (regardless of the origin of the…
I read the same thing! As someone that has followed Rossmann for years now, I am beyond disappointed that he would do something like this. I could understand if it was drug smuggling, but maple syrup? That crossed the…
“Working as intended” one might even say.
This is really interesting! Congrats on the Series A! Are there any plans to offer a simple/configurable WAF either as an included feature or as an optional add-on? EDIT: Removed one of my questions as it’s answered on…
The duality of man: I actually like them! (To be clear: I am no fan of OpenAI)
Next they’ll slowly reduce how much CC usage you can get out of the $100 Max plan, then introduce a new $300 “Max Plus” plan with “40x” usage. “You asked, and we listened: Introducing Max Plus, our biggest plan yet,…
I use Arch btw.