Files
sci-gui-agent-benchmark/run_operator.sh
2025-06-02 12:11:25 +00:00

10 lines
270 B
Bash

python run_multienv_openaicua.py \
--headless \
--observation_type screenshot \
--model computer-use-preview \
--result_dir ./results_operator_aws_new \
--test_all_meta_path evaluation_examples/test_small_debug.json \
--region us-east-1 \
--max_steps 150 \
--num_envs 5