I am currently using the Team Plan.
Is grok-code-fast-1 free to use?
On the dashboard under Usage, I see that it is being charged at $0.01.
Why don’t you offer the same free trial for GPT-5-Codex?
Dream on 
I disagree. This makes it difficult to evaluate the effectiveness of the model.
hi @furki perhaps a question to ask the model provider 
Is grok-code still free?
Dashboard is broken for old plan users (no request count), that’s why I ask @condor
I am checking it, according to the model card it is free but I want to make sure before posting an definitive answer.
God bless Elon for replacing our free Auto on time! 
Found a counter that works, grok-code still free.
Yes, zero requests.
Currently, on the billing screen, all the models have non-zero charges, that are “included”.
Sonnet 4.5 and Gpt-5 will be billed for sure. So, if i go beyond $20, I will have to pay more.
grok-code-fast-1 seems to be handled on the billing screen the same way as the other paid models.
is the billing screen wrong?
Sorry, i have no idea why. It shows “US$0.00 Included” on the latest version I’m using: Version 1.7.33 (Universal).
Does anyone know if we’ll be able to “Deny provider data collection” once this is a paid model?
Overall, I’m very impressed by how much this model can do at it’s model price and will likely pay for it if I can have data privacy.
The token accounting system is broken and it accounts for fewer tokens than users actually spend. Condor told me that everything missing from the Dashboard is not counted as a waste.
As long as the model selector says that the model is free, it is must be free.
My Cursor is also 1.7.33. The screenshots were from the billing screen on the Cursor website. I don’t know how to view the usage/billing from within Cursor IDE.
Version: 1.7.33
VSCode Version: 1.99.3
Commit: a84f941711ad680a635c8a3456002833186c4840
Date: 2025-10-03T03:28:06.574Z
Electron: 34.5.8
Chromium: 132.0.6834.210
Node.js: 20.19.1
V8: 13.2.152.41-electron.0
OS: Linux x64 6.8.0-85-generic
Grok Code is too cheap for my mental health…
The usage is only this high because it takes 3-4 prompts to get the desired results no matter how much introspection or planning you throw at it. For example this image was generated by Grok, all the bars are the same length. But at the same time, I used up all the other Included models.
Using Opus would probably be 50-75X more expensive, but at least 5X more productive.
max mode right?
No more max mode for me. But I’ll run some comparison tests.