Rendered at 22:30:22 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
biglyburrito 20 hours ago [-]
Burned through 50+% of my monthly quota today using Sonnet 4.6, after starting the day at 0%. Only used it on an average daily workload, same as what I'd done all last month, and I used maybe 52% throughout all of May.
I don't really care all that much, because my employer is the one paying for Copilot & wants us all using it. Personally, I think it sucks: the tooling is inconsistent across VSCode and Visual Studio, Microsoft caps the context window of Sonnet 4.6 at 200k even though it's capable of 1M, & at this price point there's almost certainly better ways of getting the same results from other models via other CLIs or plugins.
grandpoobah 9 hours ago [-]
Copilot for Visual Studio is just bad. How can something as simple as automatically including selected code in the query not be possible? lol. And the diffing feature always breaks. I just gave up in the end.
SomeUserName432 13 hours ago [-]
> because my employer is the one paying for Copilot & wants us all using it
I was told we're moving off Copilot and over to Claude desktop sometime this week.
Burned trough 30% of our included credits yesterday, while intentionally attempting to use less.
boesboes 13 hours ago [-]
It's like 80-100X more expensive if my calculations are correct.
Anyone surprised by this switch and bait tactic: i've been informed this is 'really smart marketing'. I haven't asked what they consider giving crack to children, but i have an idea..
cyanydeez 5 hours ago [-]
it's hyper drive enshittification; what was just a prototype business decision by nascent internet startups has been turned into a fascist defacto billionaires game. We're basically repeating the rise of the american nazis as they tried to over through the government through business conspiracy.
Cancelled my personal subscription (annual) yesterday.
- they "swallowed" my monthly subscription in January, I had to subscribe (and pay again)
- they promised tools to preview the new costs, they did too little and too badly (you have to click an export button, wait for a mail and click on a link on it, then download their csv which even showed substantial dollar costs for rows with 0 requests)
- models kept on appearing/disappearing/re-appearing-disabled on our company account in the latest weeks
- as of May 31th, I had no clue and could not tell if I would been migrated to token billing, or would have to stay with the moronic new multipliers. News came on Jun 1st, of course
They don't really look able or willing to properly manage their own product at the moment. And yes, new subscription are paused, so I won't be able to re-subscribe.
Quite frankly the only reason to go copilot is to have it in the VSCode chat (and yes, there's some chance to use it BYOK, provided it works).
Besides, their offering even at market prices looks inferior to what you can get elsewhere. You can use DeepSeek and pay pennies, use Fireworks and have the choice to use cheaper open models (which GitHub does not provide, and are actually good and even better than Claude sometimes), or subscribe to Open Router and use virtually anything.
I still have no idea if cancelling my subscription will get any money back, probably not.
mgrund 16 hours ago [-]
The big question is if they manage to keep on their enterprise consumers. Vowing changes nothing, they will be looking at the actual numbers.
Dissatisfaction is kind of expected (my cost goes up 2 orders of magnitude and I already cancelled since there are better options at market rate). Complaining without change won’t matter.
maverwa 6 hours ago [-]
With big enterprises, at least the one my account hangs on, the process of replacing the supplier would take months just to get through all the enterprise hoops. Not sure if anyone wants to get on that. I would assume buying Microsoft is somewhat simpler than buying some startup not yet on the holy list of divine supplies. Said enterprise just this year managed to bulk buy JetBrains licenses for everyone that then got suspended. For everyone. Twice. With 30days between cases. So I may be cynic but that’s earned!
janstice 16 hours ago [-]
This the the pricing model for Anthropic now too for business customers over 150 seats - under 150 you can take the Team plan at $25/mth & it’s similar to the personal plans, but Enterprise you’re billed by the token (with a $20/seat minimum but the tokens are pooled together). This is the AI Uber moment where the VCs stop subsiding your fares.
pqtyw 12 hours ago [-]
To be fair the margins on inference should be pretty good for Anthropic and OpenAI.
The token cost for very large open models seems to indicate that. Of course DeepSeek is probably subsidizing their current pricing to a significant extent but then you have all other provides on OpenRouter and I can't imagine it would make sense for them to operate at a significant loss since they don't offer anything but inference.
Anthropic is just utilizing its position to maximize the prices as much as possible and see what the market can bear and of course they are not even close to offsetting their R&D and capital investments.
If LLMs become commoditized and open models become good enough for most use cases its hard to imagine that prices won't drop (and the AI bubble popping as a consequence of that...)
baobabKoodaa 12 hours ago [-]
What are the better options? I need something that works in VSCode.
Not only did they switch to token billing and increase the multipliers, they started auto selecting 6x and 9x models more often than before. The UX is such that you don’t find out what model it chose until after the request is complete.
bob1029 16 hours ago [-]
I've been looking at the warning banner in VS2026 for the last few weeks with total indifference. Prepaid tokens tend to go a long way when you pay for them with your own money.
I suspect most of the developers in the angry set have been using Ralph loops while they sleep or are being ran in circles all day with aimless prompting (bad leadership).
If you know what you want done, are being encouraged to go do it, and have some experience regarding the space of possible outcomes, you can go further on one token than your peers can on a thousand.
general1465 15 hours ago [-]
> If you know what you want done, are being encouraged to go do it, and have some experience regarding the space of possible outcomes, you can go further on one token than your peers can on a thousand.
Tortoise and hare story.
throw03172019 20 hours ago [-]
Why would you even use GitHub Copilot. It’s so far behind. They kind of blew it.
pjmlp 17 hours ago [-]
Company policies, a bit like having teams.
baobabKoodaa 12 hours ago [-]
What are good alternatives that work in VSCode?
Natfan 18 hours ago [-]
using the subscription with a third party harness?
baobabKoodaa 12 hours ago [-]
Can someone recommend good alternatives that work in VSCode? I just want autocomplete, I don't want any agent bs.
menno-sh 9 hours ago [-]
afaik autocomplete is still included
baobabKoodaa 8 hours ago [-]
included in what?
fancyfredbot 5 hours ago [-]
If you pay for any tier of copilot then you will get unlimited access to the auto complete.
If you don't use agents it's possible you'll not notice much of a change to copilot
baobabKoodaa 4 hours ago [-]
I cancelled my autopilot subscription because Microsoft communicated to me that I would have to pay metered use in addition to the flat rate which I already pay. You're telling me this was misinfo? Microsoft actually is not starting to meter usage of autocomplete by GitHub Copilot? Why did they send this message then?
I don't really care all that much, because my employer is the one paying for Copilot & wants us all using it. Personally, I think it sucks: the tooling is inconsistent across VSCode and Visual Studio, Microsoft caps the context window of Sonnet 4.6 at 200k even though it's capable of 1M, & at this price point there's almost certainly better ways of getting the same results from other models via other CLIs or plugins.
I was told we're moving off Copilot and over to Claude desktop sometime this week.
Burned trough 30% of our included credits yesterday, while intentionally attempting to use less.
https://en.wikipedia.org/wiki/Business_Plot
- they "swallowed" my monthly subscription in January, I had to subscribe (and pay again)
- they promised tools to preview the new costs, they did too little and too badly (you have to click an export button, wait for a mail and click on a link on it, then download their csv which even showed substantial dollar costs for rows with 0 requests)
- models kept on appearing/disappearing/re-appearing-disabled on our company account in the latest weeks
- as of May 31th, I had no clue and could not tell if I would been migrated to token billing, or would have to stay with the moronic new multipliers. News came on Jun 1st, of course
They don't really look able or willing to properly manage their own product at the moment. And yes, new subscription are paused, so I won't be able to re-subscribe.
Quite frankly the only reason to go copilot is to have it in the VSCode chat (and yes, there's some chance to use it BYOK, provided it works).
Besides, their offering even at market prices looks inferior to what you can get elsewhere. You can use DeepSeek and pay pennies, use Fireworks and have the choice to use cheaper open models (which GitHub does not provide, and are actually good and even better than Claude sometimes), or subscribe to Open Router and use virtually anything.
I still have no idea if cancelling my subscription will get any money back, probably not.
Dissatisfaction is kind of expected (my cost goes up 2 orders of magnitude and I already cancelled since there are better options at market rate). Complaining without change won’t matter.
The token cost for very large open models seems to indicate that. Of course DeepSeek is probably subsidizing their current pricing to a significant extent but then you have all other provides on OpenRouter and I can't imagine it would make sense for them to operate at a significant loss since they don't offer anything but inference.
Anthropic is just utilizing its position to maximize the prices as much as possible and see what the market can bear and of course they are not even close to offsetting their R&D and capital investments.
If LLMs become commoditized and open models become good enough for most use cases its hard to imagine that prices won't drop (and the AI bubble popping as a consequence of that...)
I suspect most of the developers in the angry set have been using Ralph loops while they sleep or are being ran in circles all day with aimless prompting (bad leadership).
If you know what you want done, are being encouraged to go do it, and have some experience regarding the space of possible outcomes, you can go further on one token than your peers can on a thousand.
Tortoise and hare story.
If you don't use agents it's possible you'll not notice much of a change to copilot