Welcome to free-claude-code Discussions!

[Alishahryar1]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

[poeta-liado]

Hey all 🫶🏻

[pri4350]

hey

[aulite]

hey

[belingham076]

hey

[Se7en0512]

Free-claude-code for VS code, how?

[abindacarmo]

its free version can help what else level for coding? it is ready enough for mini project like simple website?

[poeta-liado]

I just started coding 3 days ago and this website has been my saviour
Learning AI coding is the way forward.

[Libonatti93]

Hello!

[Kashan-2912]

hello guys

[moha-elh]

i followed all the steps in the readme file, but i still got the issue that claude command isn't found?!

[mymel2001-holder]

You have to install Clade Code on the target machine first. (The client.) This isn't a Claude Code like client, but a server to act as a proxy so you can use stuff like NVIDIA NIM or Ollama models to effectively get "Claude Code" for free.

[moha-elh]

isn't claude code gives u the ability to plug in external api by default?!

[Alishahryar1]

isn't claude code gives u the ability to plug in external api by default?!

Not openai compatible ones

[999prithviii]

Yo Ali, how do you input the API Keys after doing the cloning process im a little lost need a bit of help

[mymel2001-holder]

Hello, world! My name is Sammy, but you can just call me "Mel." I am a vibe coder who loves local systems, FOSS, but most of all LARPing as a normie by using Apple as my ecosystem of choice instead of something like... say... Linux.

It's about 2am for me rn. I really should be going to bed, but there's this issue thats bugging me, I guess I will talk check for a relevant thread and talk about it there.

[mymel2001-holder]

update: noticed the issues i was having were already reported (the ones due to qwen3.6 locally and deepseek v4 pro on nvidia nim)

[999prithviii]

Im having an issue installing and running the AI
Im New to all this can you explain how you input the API Key after the cloning process?

[Sathikumar14]

iam very glad to know that

[eZ4RK0]

Hello, is it possible to replace the OpenRouter API URL with one of my own? I have a self-hosted AI server with my own proxy (not Ollama, but a completely homemade proxy) and I’m using an API similar to OpenRouter ?

[iandevenv-ops]

Hello. I am currently very happy to finally use Claude Code and for FREE!!! I've been using NVIDIA nim to power it for a few days, but for long tasks, the model kind of stops midway with the frustrating request timeout or other outrageous errors... can someone tell me why?

[vaibhavingle2007]

now this repo is not that useful as the launch of openclaude we can use any open compatible models there also local so no need of this repo as it take much memory in background for calling the api then sending it to claude code every one should try that

[999prithviii]

Hey Guys!!
Im new here.

I'm trying to set up the NVIDIA NIM Provider but i dont know how to paste in the api key and how to execute it
I Have my API key but i dont know how to choose my provider and how to input the API Key.

Can anyone with setup experience help?

[axgoodhart]

If you do not know this it is a long path

[ukeepit]

Just find the .env file and paste your API key at the top of it in which the variable name is 'NVIDIA_NIM_API_KEY'.After finishing this and confirming you can chat with claude freely then to consider how to change another provider.

[danatzh]

hey guys! could yall help me out with this problem?
I'm using deepseek api tho :(

[RR-J15]

the problem is when i use the product claude often respond failedrequest like the request take too much time , and some of the models in the .env files are not free actually ; @Alishahryar1

[RonalMZ]

¿Alguien interesado en formar un grupo de trabajo?

[Susanthakuri92]

Only opus works for me, other models doesnt work that is sonnet and haiku i get provider api request failed.
i tried changing the models but got the same error

[poeta-liado]

Hola,Wn que tipo de y estas pensando? Sent from my iPhoneOn 29 Apr 2026, at 17:26, Ronal_Angulo ***@***.***> wrote: ¿Alguien interesado en formar un grupo de trabajo? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you commented.Message ID: ***@***.***>

[ckckckckckckckeaaa]

add codex oauth please

[Akwolupeter]

How can I host my database on remote server?

[IMPULSINATOR]

Did anyone actually successfully use it with ollama? im using gemma 4 e4b and am unable to use this with that.

[mymel2001-holder]

I was able to use some models (I think qwen3.5:9b), but not most newer ones.

[mdobns]

Did anyone actually successfully use it with ollama? im using gemma 4 e4b and am unable to use this with that.

Here are few cloud models which currently working with free api keys

Ollama

MODEL_OPUS="ollama/gpt-oss:120b-cloud"
MODEL_SONNET="ollama/qwen3-coder-next:cloud"
MODEL_HAIKU="ollama/minimax-m2.5:cloud"

Nvdia Nim

MODEL_OPUS="nvidia_nim/mistralai/mistral-nemotron"
MODEL_SONNET="nvidia_nim/qwen/qwen3-coder-480b-a35b-instruct"
MODEL_HAIKU="nvidia_nim/minimaxai/minimax-m2.7"