While trying locally with env="playwright" it also hits gemini-2.5-pro

[drashyakuruwa]

Hello,

I am not sure if I'm facing an issue or if this is the expected behaviour.

I was setting up the project on my local Windows 11 Pro machine and ran the sample command: python main.py --query="Go to Google and type 'Hello World' into the search bar" --env="playwright"

Initially, I received an "API key limit exhausted" error while using a free-tier Gemini API key. I then created a paid-tier key and retried the command. This time it worked, but the execution was a bit slow.

I tried a different query, and while it also executed successfully, the performance was still slow.

When I checked my API key usage, I saw the following:

When I checked the code in the main.py and agent.py files, I could only find this configuration.:

parser.add_argument(
    "--model",
    default='gemini-2.5-computer-use-preview-10-2025',
    help="Set which main model to use.",
)

Why is gemini-2.5-pro also being used, and is this the expected behaviour? If it is, can I change the configuration to use a faster model like gemini-2.5-flash?

[NJX-njx]

First, I suggest you check if gemini-2.5-pro was used for handling other things on that day. Then, the model gemini-2.5-computer-use-preview-10-2025 should be a custom one by you. Maybe it is still based on gemini-2.5-pro at the bottom? Finally, regarding the gemini-2.5-flash model, I suggest you check if its available functions correspond to those in the project.