3 ms·
How does it compare to this posted less than 24 hours ago? https://news.ycombinator.com/item?id=39698546 https://news.ycombinator.com/item?id=39698546
by spxneo 3y ago
How does it compare to this posted less than 24 hours ago?
https://news.ycombinator.com/item?id=39698546 https://news.ycombinator.com/item?id=39698546
- suchintan 3y agoSaw the launch yesteday. Love all of the excitement in the space! LaVague is all about generating selenium code to interact with a specific page, and do it step-by-step Skyvern is all about taking a simple instruction and converting it to a series of LLM-driven actions. It's meant to be more autonomous ("tell Skyvern what to do")
- spxneo 3y agoIsn't that the same thing when you interact with the underlying webpage?
- suchintan 3y agoWe're quite different than LaVague. LaVague passes in the entire HTML DOM to the LLM to help it generate XPaths and valid Selenium code. (https://github.com/lavague-ai/LaVague/blob/main/src/lavague/default_prompt.py#L203 https://github.com/lavague-ai/LaVague/blob/main/src/lavague/...) Try this at your own risk.. any reasonable website would result in extraordinarily high input token costs We spend quite a bit of our time building a layer between the HTML and the LLM call to distill important pieces of information down to actions the LLM can take.. better weighing cost vs output. We're still not at 100% coverage.
- LZ_Khan 3y agoIt is similar. hence the timing of the plug, probably :)