diff options
Diffstat (limited to 'benchmark/reports/mini-agi/8.1_TestBasicMemory.json')
-rw-r--r-- | benchmark/reports/mini-agi/8.1_TestBasicMemory.json | 30 |
1 files changed, 30 insertions, 0 deletions
diff --git a/benchmark/reports/mini-agi/8.1_TestBasicMemory.json b/benchmark/reports/mini-agi/8.1_TestBasicMemory.json new file mode 100644 index 000000000..66cc2f9ae --- /dev/null +++ b/benchmark/reports/mini-agi/8.1_TestBasicMemory.json @@ -0,0 +1,30 @@ +{ + "command": "agbenchmark start --test TestBasicMemory", + "completion_time": "2023-07-17-18:22", + "metrics": { + "run_time": "53.48 seconds", + "highest_difficulty": "basic: 2" + }, + "tests": { + "TestBasicMemory": { + "data_path": "agbenchmark/challenges/memory/m1", + "is_regression": false, + "task": "Follow the instructions in the instructions_1.txt file", + "answer": "2314", + "description": "Tests ability for the agent to remember information between each action. An id is presented initially and the agent has to remember it after reading 4 other files", + "metrics": { + "difficulty": "basic", + "success": true, + "success_%": 75.0, + "run_time": "53.252 seconds" + }, + "reached_cutoff": false + } + }, + "config": { + "workspace": "${os.path.join(Path.home(), 'miniagi')}" + }, + "additional": { + "model": "gpt-4" + } +} |