diff options
Diffstat (limited to 'benchmark/agbenchmark/challenges/verticals/scrape/3_revenue_retrieval/data.json')
-rw-r--r-- | benchmark/agbenchmark/challenges/verticals/scrape/3_revenue_retrieval/data.json | 31 |
1 files changed, 31 insertions, 0 deletions
diff --git a/benchmark/agbenchmark/challenges/verticals/scrape/3_revenue_retrieval/data.json b/benchmark/agbenchmark/challenges/verticals/scrape/3_revenue_retrieval/data.json new file mode 100644 index 000000000..f76016405 --- /dev/null +++ b/benchmark/agbenchmark/challenges/verticals/scrape/3_revenue_retrieval/data.json @@ -0,0 +1,31 @@ +{ + "category": [ + "scrape_synthesize", + "general" + ], + "cutoff": 60, + "dependencies": [ + "TestBasicRetrieval" + ], + "eval_id": "dc2114d7-1597-4c9b-bed0-a97937ad977f", + "ground": { + "answer": "It was $81.462 billion in 2022. In millions the answer is 81,462.", + "eval": { + "type": "file" + }, + "files": [ + ".txt" + ], + "should_contain": [ + "81,462" + ], + "should_not_contain": [] + }, + "info": { + "description": "Tests if the agent can retrieve Tesla's revenue in 2022.", + "difficulty": "intermediate", + "side_effects": [] + }, + "name": "RevenueRetrieval", + "task": "Write tesla's exact revenue in 2022 into a .txt file. Use the US notation, with a precision rounded to the nearest million dollars (for instance, $31,578 million)." +} |