How to use this model?
I'm looking for the Local LLM as replacement for expensive Cloud LLM (Anthropic, OpenAI) for Navigating tasks as well as performing the tasks similar as Computer Use, Operator, etc.
Does this replace above Cloud based models for similar Agents activities?
I'm interesting in this in case my guess is correct,
Thanks
It may be able to act as a replacement for some of those tasks.
It is fine-tuned specifically to use tools from a playwright MCP server (automated web browsing).
If you want to use it as a web agent it hopefully will perform well, I am still working on this and haven't evaluated it's performance yet.
It will probably not work well for other desktop tasks.
It may be able to act as a replacement for some of those tasks.
It is fine-tuned specifically to use tools from a playwright MCP server (automated web browsing).
If you want to use it as a web agent it hopefully will perform well, I am still working on this and haven't evaluated it's performance yet.It will probably not work well for other desktop tasks.
Can you share some guidances on how to use this fine-tune model (for automated web browsing in details) as well as the procedure on how to fine-tune it (inputs, tool/technique (Unsloth, I guess), and outcome)?
Yes I will share them when I have time, you can look at my playwright-mcp dataset I have a lot of my files stored there
FYI I just added most of the files necessary and updated the README
Thanks a lot for your info,
How to use this trained model for inferencing while calling Playwright-MCP?
I'm interested in score card of this compare with Cloud AI API such as OpenAI, Anthropic (I use this most of the times with Playwright-mcp)