[Bug]: Workers AI broken

Is there an existing issue for the same bug?

• I have searched existing issues and this is not a duplicate.

Bug Description

I'm trying to use @hf/thebloke/codellama-7b-instruct-awqbut it's saying The configured model has a context window of 4,096 tokens, which is below the minimum of 16,384 tokens required for OpenHands to function properly. For local LLMs (Ollama, LM Studio, etc.), increase the context window. For cloud providers, verify you're using the correct model variant. For configuration instructions, see: https://docs.openhands.dev/openhands/usage/llms/local-llms To override this check (not recommended), set the environment variable: ALLOW_SHORT_CONTEXT_WINDOWS=true
Even though I have configured it right

Expected Behavior

No response

Actual Behavior

No response

Steps to Reproduce

No response

OpenHands Installation Method

OpenHands Cloud (app.all-hands.dev)

If you selected "Other", please specify

No response

OpenHands Version

Web

Version Confirmation

• I have confirmed this bug exists on the LATEST version of OpenHands

Model Name

@hf/thebloke/codellama-7b-instruct-awq

Operating System

Other

Browser (if using web UI)

Safarl

Logs and Error Messages

litellm.AuthenticationError: AuthenticationError: OpenAIException - {
"error": {
"message": "Incorrect API key provided: cfut_SLa*****************************************335a. You can find your API key at https://platform.openai.com/account/api-keys.",
"type": "invalid_request_error",
"param": null,
"code": "invalid_api_key"
}
}

Screenshots and Additional Context

No source file, test, or entry point is identified. Start by reproducing the issue in OpenHands Cloud with the configured @hf/thebloke/codellama-7b-instruct-awq model, then compare the context-window and API-key errors with the provider configuration. Done means the configured model is accepted and requests no longer fail with these errors.