Maybe one day Claude can rewrite its interface to be more accessible to blind people like me.
Claude now has access to a server-side container environment
21–30 of 363 posts
Re: Claude now has access to a server-side container environment
#22They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
The people shipping these features are not the same people who are fixing reliability probably.
No, but the salaries of the people shipping those fixtures could be spent on people who can fix the reliability problems.
Re: Claude now has access to a server-side container environment
#23They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
Some of this has gotta be people asking more of it than they did before, and some has gotta be people who happened to use it for things it's good at to begin with and are now asking it things it's bad at (not necessarily harder things, just harder for the model).
However there have been some bugs causing performance degradation acknowledged by Anthropic as well (and fixed) and so I would guess there's a good amount of real degradation still if people are still seeing issues.
I've seen a lot of people switching to codex cli, and yesterday I did too, for now my 200/mo goes to OpenAI. It's quite good and I recommend it.
Re: Claude now has access to a server-side container environment
#24They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
Anthropic claims that they don't degrade models under load, and the performance issues were a result of a system error: https://status.anthropic.com/incidents/72f99lh1cj2c That being said, they still have capacity issues on any day of the week that ends in Y. No clue how long would that take to resolve.
Frankly, I don't believe their claims that they don't degrade the models. I know we see models as less intelligent as we get used to them and their novelty wears off but I've had to entirely give up on Claude as a coding assistant because it seems to be incapable of following instructions anymore.
Re: Claude now has access to a server-side container environment
#25Earlier quoted context omitted.
The people shipping these features are not the same people who are fixing reliability probably.
No, but the salaries of the people shipping those fixtures could be spent on people who can fix the reliability problems.
Wait—surely they're dogfooding Claude for infrastructure tasks, making existing engineers 10x as productive, requiring less human engineers overall?
Re: Claude now has access to a server-side container environment
#26They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
Anthropic claims that they don't degrade models under load, and the performance issues were a result of a system error: https://status.anthropic.com/incidents/72f99lh1cj2c That being said, they still have capacity issues on any day of the week that ends in Y. No clue how long would that take to resolve.
Not nitpicking, but they said:
> we never intentionally degrade model quality as a result of demand or other factors
Fully giving them the benefit of the doubt, I still think that still allows for a scenario like "we may [switch to quantized models|tune parameters], but our internal testing showed that these interventions didn't materially affect end user experience".
I hate to parse their words in this way, because I don't know how they could have phrased it that closed the door on this concern, but all the anecdata (personal and otherwise) suggests something is happening.
Re: Claude now has access to a server-side container environment
#27It does this in emacs with efrit. https://github.com/steveyegge/efrit
It can actually drive emacs itself, creating buffers, being told not to edit the buffers and simply respond in the chat etc.
I actually _like_ working with efrit vs other LLM integrations in editors.
In fact I kind of need to have my anthropic console up to watch my usage... whoops!
Re: Claude now has access to a server-side container environment
#28They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
This can't be understated. I started using it heavily earlier this summer and it felt like magic. Someone signing up now based on how I described my personal experiences with it then would think I was out of my mind . For technical tasks it has been a net negative for me for the last several weeks. (Speaking of both Claude Code and the desktop app, both Sonnet and Opus >=4, on the Max plan.)
I hesitate to use phrases like "bait and switch" but it seems like every model gets released and is borderline awe-inspiring, then as adoption increases, and load increases, it's like it gets hit in the head with a hammer and is basically useless for anything beyond a multi-step google search.
Re: Claude now has access to a server-side container environment
#29They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
At least some transparency would be nice. It feels like they are serving less intelligent models labelled as more intelligent ones during peak times.
Re: Claude now has access to a server-side container environment
#30They need to focus on fixing reliability first. Their systems constantly go down and it appears they are having to quantise the models to keep up with demand, reducing intelligence significantly. New features like this feel pointless when the underlying model is becoming unusable.
This can't be understated. I started using it heavily earlier this summer and it felt like magic. Someone signing up now based on how I described my personal experiences with it then would think I was out of my mind . For technical tasks it has been a net negative for me for the last several weeks. (Speaking of both Claude Code and the desktop app, both Sonnet and Opus >=4, on the Max plan.)
Have you considered perhaps that you are, indeed, out of your mind? Or more precisely, that you could be rationalizing what is essentially a random process?
Based on the discussions here it seems that every model is either about to be great or was great in the past but now is not. Sucks for those of us who are stuck in the now, though.