diff options
Diffstat (limited to 'benchmark/reports/mini-agi/5_TestRevenueRetrieval_1.0.json')
-rw-r--r-- | benchmark/reports/mini-agi/5_TestRevenueRetrieval_1.0.json | 29 |
1 files changed, 29 insertions, 0 deletions
diff --git a/benchmark/reports/mini-agi/5_TestRevenueRetrieval_1.0.json b/benchmark/reports/mini-agi/5_TestRevenueRetrieval_1.0.json new file mode 100644 index 000000000..28d091d28 --- /dev/null +++ b/benchmark/reports/mini-agi/5_TestRevenueRetrieval_1.0.json @@ -0,0 +1,29 @@ +{ + "command": "agbenchmark start --test TestRetrieval2", + "completion_time": "2023-07-17-13:54", + "metrics": { + "run_time": "36 seconds", + "highest_difficulty": "TestRetrieval2: 3" + }, + "tests": { + "TestRetrieval2": { + "data_path": "agbenchmark/challenges/retrieval/r2_tesla_revenue", + "is_regression": false, + "reached_cutoff": false, + "answer": "It was $81.462 billion in 2022.", + "description": "A no guardrails search for info", + "metrics": { + "difficulty": "novice", + "success": true, + "success_%": 50.0, + "run_time": "35.59 seconds" + } + } + }, + "config": { + "workspace": "${os.path.join(Path.home(), 'miniagi')}" + }, + "additional": { + "model": "gpt-4" + } +} |