Error code: 400

[NShravanReddy]

Hi,

Here's my full code repo https://github.com/NShravanReddy/secure-computer-use/tree/os-computer-use

To reproduce the error I am getting,

1. I cloned the repo and then installed mini forge
2. Modified the llm.py file code in 16, 17 lines as,

LLAMA_32_OPENROUTER_MODEL = "gemini-2.0-flash-exp"
LLAMA_33_FIREWORKS_MODEL = "accounts/fireworks/models/llama-v3p3-70b-instruct"

3. and 49 to 55 lines as,

fireworks = OpenAI (
base_url="https://generativelanguage.googleapis.com/v1beta/openai/",
api_key=os-getenv("FIREWORKS_API_KEY"),
)
llama_vision = OpenAI(
base_url="https://generativelanguage.googleapis.com/vlbeta/openai/", 
api_key=os-getenv ("'OPENROUTER_API_KEY"),
)
osatlas = Client (OSATLAS_HUGGINGFACE_SOURCE)

4. Then I [https://github.com/NShravanReddy/secure-computer-use/tree/os-computer-use?tab=readme-ov-file#get-started] followed step 2 to 4 mentioned in the official repo
5. In the .env file I have given api keys

E2B_API_KEY="my e2b api key"
OPENROUTER_API_KEY="open router api key"
FIREWORKS_API_KEY="fireworks api key"

I am getting the below error
e2b error.txt

[jamesmurdza]

Hi,

Can you please introduce the change you're trying to make? It looks like you're trying to replace one of the models with Gemini.

Please note: We're using three different models in this agent, so just changing one of them may not have the result you intend.

James

[NShravanReddy]

Hi ,

Change is to use gemini model