mirror of
https://github.com/aljazceru/Auto-GPT.git
synced 2026-01-03 22:34:29 +01:00
51 lines
1.9 KiB
JSON
51 lines
1.9 KiB
JSON
{
|
|
"command": "agbenchmark start --maintain --mock",
|
|
"completion_time": "2023-08-02-00:07",
|
|
"benchmark_start_time": "2023-08-02-00:07",
|
|
"metrics": {
|
|
"run_time": "0.6 seconds",
|
|
"highest_difficulty": "basic: 2"
|
|
},
|
|
"tests": {
|
|
"TestPlanCreation": {
|
|
"data_path": "agbenchmark/challenges/content_gen/2_plan/data.json",
|
|
"is_regression": true,
|
|
"category": [
|
|
"content_gen"
|
|
],
|
|
"task": "As an agent powered by a large language model, create a plan for yourself to execute on based on the task of booking a flight to Toronto. Output this plan into a .txt file.",
|
|
"answer": "1. Search for flights to Toronto.\n2. Select the cheapest flight.\n3. Book the flight.\n4. Pay for the flight.\n5. Confirm the flight.",
|
|
"description": "Tests ability to generate content based on the content of 2 files.",
|
|
"metrics": {
|
|
"difficulty": "basic",
|
|
"success": true,
|
|
"success_%": 100.0,
|
|
"cost": null,
|
|
"run_time": "0.003 seconds"
|
|
},
|
|
"reached_cutoff": false
|
|
},
|
|
"TestWriteFile": {
|
|
"data_path": "agbenchmark/challenges/interface/write_file/data.json",
|
|
"is_regression": true,
|
|
"category": [
|
|
"interface"
|
|
],
|
|
"task": "Print the the capital of America to a .txt file",
|
|
"answer": "Washington",
|
|
"description": "Tests the writing to file",
|
|
"metrics": {
|
|
"difficulty": "interface",
|
|
"success": true,
|
|
"success_%": 100.0,
|
|
"cost": null,
|
|
"run_time": "0.002 seconds"
|
|
},
|
|
"reached_cutoff": false
|
|
}
|
|
},
|
|
"config": {
|
|
"workspace": "generated",
|
|
"entry_path": "agbenchmark.benchmarks"
|
|
}
|
|
} |