Sign in to join the discussion — post, comment, and like with a magic link or social login.

Sign In
Adadi

Adadi

i tried running claude code for free in vscode after following your video but after all the steps you mention i tried but keep having this error in claude code creditd too low. will appreciate if you can help
8 comments

Sign in to join the discussion.

Hasan
Hasan
are you using claude in terminal or the extension ?
which model you are using?
Adadi
Adadi
im using the claude extension in vscode i copied the code you gave in the video description and insert the openrouter api key i generated
i creatde the .claude folder and the local.settings.json file and copy and paste the paste the code you provided and inserted just the api key i generated from openrouter
Hasan
Hasan
@Adadi based on my youtube video, it will work in terminal, not extention.
Adadi
Adadi
@Hasan sorry don't know if im missing something after typing claude as you did in the vidoe at the ternimal i can not see the model below is what i get:
PS C:\Users\pndag\.claude\.claude> claude
Welcome to Claude Code v2.1.220

Let's get started.

Choose the text style that looks best with your terminal
To change this later, run /theme

1. Auto (match terminal)
Welcome to Claude Code v2.1.220

Claude Code can be used with your Claude subscription or billed based on API usage through your Console account.

Select login method:

❯ 1. Claude account with subscription · Pro, Max, Team, or Enterprise
2. Anthropic Console account · API usage billing
3. 3rd-party platform · Amazon Bedrock, Microsoft Foundry, or Vertex AI
Hasan
Hasan
@Adadi select 2, first time, without charging, just generate an API key. And then you can access the app
Cynthia Diane
Cynthia Diane
@Hasan I'm actually using Claude code with Ollama Cloud models. Do you know if there is any way around the maximum allowed token for context window? For Deep Seek V4 Pro, it's only allowing me 200k And my sessions are starting with half the context because of the size of my workspace.
Hasan
Hasan
context window depends on the model you are using.
but loading 100K in one shot on load means it is reading too much file on load. what kind of project you are working on ?
Cynthia Diane
Cynthia Diane
@Hasan Yes, I keep running into that issue and the weird part is when using Opencode I do not have this context issue.
Deepseek is 1M context window but claude code is having trouble recognizing it. I saw Anthropic mentioned some issues around it.

For the longest time I was thinking of having different workspaces, but this is also not a solution for my work (I create marketing tools and assessments + websites and conduct a lot of research to create SOP's and handbooks for AI agents).
I'm digging into some extra information about file structure and will conduct an audit again to see how I can better optimize.

I have also decided to start testing jcode instead. A lot of the issues I have with claude code when trying to use it with something other than anthropic is not worth it, especially for non tech people I feel they are making it too complicated where other harnesses don't.

Any insight is welcome haha