Rendered at 00:56:50 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
lbreakjai 1 hours ago [-]
I cancelled my subscription two days ago. I was a customer since October last year. I could get a decent bit of work done on just the 20$ subscription, but since this monday, I can barely get two prompts in before hitting my limit.
Same codebase, same sort of prompt, same scale. I was already on the fence. Models like Qwen, Kimi, or GLM5 already go a very long way while being vastly cheaper, and the new openAI models feels equivalent but with higher limits.
This is getting to the point where the right harness makes a bigger difference than the right model. I've been experimenting with some planner-executor-reviewer setup in opencode, and I'm starting to feel like multiple smaller models working together are netting me better results.
_the_inflator 2 hours ago [-]
With the introduction of Opus 4.6 my bills went through the roof. I never burned budgets so fast with so few prompts since then.
I more and more use Codex, because token usage is a blackbox and I think that we will see the next couple of month the usual three tier model evolving: free, normal, luxury.
2027 will be the year of token regulation by administrations worldwide. Until then take care for being ripped of at the luxury level.
sunnybeetroot 3 hours ago [-]
I love the interface of Claude Code but with these limits I’d be willing to use Codex. Anyone know if it’s possible to use Claude Code with other provider subscriptions (not API usage costs)?
HDBaseT 2 hours ago [-]
You can!
According to the GLM documentation at least, you can edit you .env (~/.claude/settings.json) with an API key from the GLM Settings page. [0].
Whilst you need to generate an API Key, you are still using the 'GLM Coding Plan' as per 'Methods for Using the GLM Coding Plan in Claude Code' documentation.
Anthropic is probably the only AI company that is trying to stop absolutely hemorrhaging money and reach profitability and financial sustainability.
They aren't far off - they burn a tiny fraction of the cash of OAI and achieve similar ARR despite this - but as they tighten the belt it's inevitable that companies like OAI come in and offer more subsidized (unsustainable) inference to get people to switch. They will inevitably do the same "rug pull".
It'll be interesting to see how this plays out.
arctic-true 46 minutes ago [-]
The winner will be the last man standing. The company which is able to continue subsidizing tokens after all others have been forced to throttle and raise prices. Soak up the market share, wait for your rivals’ investors to get cold feet, then raise prices the morning after the last one folds. (It’s Google it’s always been Google)
cyanydeez 3 hours ago [-]
This will just get worse, particularly when you consider just how shitty the US government is becoming at properly managing the basic necessities for stability.
IF you arn't planning a local LLm strategy, you're surely tying your lifeline to anchors.
Same codebase, same sort of prompt, same scale. I was already on the fence. Models like Qwen, Kimi, or GLM5 already go a very long way while being vastly cheaper, and the new openAI models feels equivalent but with higher limits.
This is getting to the point where the right harness makes a bigger difference than the right model. I've been experimenting with some planner-executor-reviewer setup in opencode, and I'm starting to feel like multiple smaller models working together are netting me better results.
I more and more use Codex, because token usage is a blackbox and I think that we will see the next couple of month the usual three tier model evolving: free, normal, luxury.
2027 will be the year of token regulation by administrations worldwide. Until then take care for being ripped of at the luxury level.
According to the GLM documentation at least, you can edit you .env (~/.claude/settings.json) with an API key from the GLM Settings page. [0].
{ "env": { "ANTHROPIC_AUTH_TOKEN": "your_zai_api_key", "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic", "API_TIMEOUT_MS": "3000000" } }
Whilst you need to generate an API Key, you are still using the 'GLM Coding Plan' as per 'Methods for Using the GLM Coding Plan in Claude Code' documentation.
[0] - https://docs.z.ai/devpack/tool/claude
They aren't far off - they burn a tiny fraction of the cash of OAI and achieve similar ARR despite this - but as they tighten the belt it's inevitable that companies like OAI come in and offer more subsidized (unsustainable) inference to get people to switch. They will inevitably do the same "rug pull".
It'll be interesting to see how this plays out.
IF you arn't planning a local LLm strategy, you're surely tying your lifeline to anchors.