-
Notifications
You must be signed in to change notification settings - Fork 2.1k
Now get's list of models from ollama instead of hardcoding them. also Improve Browser navigation and error handling #331
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
|
- Add isolation of navigation actions to ensure page loads completely before further actions - Add browser context validation to catch initialization errors early - Add state validation to ensure browser state is available before proceeding The navigation isolation particularly helps with Ollama models where sequential actions might be attempted before page load completes. Browser context and state validation provides earlier, clearer error messages for common failure cases.
Mainly for making Ollama more reliable but probably others as well. |
Modified the llm_num_ctl to be more approriate for various models Updated llm so models are consolidated fixed some errors in utils.py
This is much more stable At this point using ollama I am able to run the default agent task (opening openai and clicking the first link). It's also fairly fast. I've not tested if any of these changes have affected the other models so that needs to be tested, but I'm hopeful that they will work as before or better. |
hi, Please don't change so much content in one pull request. It's very difficult for me to review and verify. Also, please don't change content that doesn't need to be modified for now. |
You should be able to do three separate pull requests. (I don't know why they were all concatenated). |
@iplayfast great pr, but you only opened one PR. |
Fix for issue #330
Dynamically loads models from Ollama.