From the text_to_sql example: [0]<p>My question: "Which waiter got second most total money from tips? create a jpeg with a bar chart of the total tips for each waiter."<p>Step 0 included a correct sql query.
Step 1 included python code with matplotlib, which failed because matplotlib was not in the allowed imports. The system then pivoted to another python script which printed a bar plot with ## characters.<p>Final answer: The waiter with the second most total tips is Corey Johnson<p>Bar Chart of Total Tips for Each Waiter:<p>Michael Watts ################################################## (5.67)<p>Corey Johnson ########## (1.2)<p>Margaret James ######## (1.0)<p>[0] - <a href="https://github.com/huggingface/smolagents/blob/main/docs/source/en/examples/text_to_sql.md">https://github.com/huggingface/smolagents/blob/main/docs/sou...</a>
Congrats on the launch. I actually really like how the agency level x capability is displayed on the page. Too many text in / text out abstractions being mislabeled as “agents” these days.