A framework to enable multimodal models to operate a computer.
COMMITS
/ operate/operate.py February 17, 2024
J
Improve app `print` experience
Josh Bickett committed
February 9, 2024
M
Add `--verbose` flag and directly access verbose flag from Config singleton
Michael Hogue committed
February 2, 2024
J
remove extra space in `print`
Josh Bickett committed
January 21, 2024
J
Add `SYSTEM_PROMPT_OCR_MAC` and `SYSTEM_PROMPT_OCR_WIN_LINUX`
Josh Bickett committed
January 19, 2024
J
remove `print`
Josh Bickett committed
J
Add `initialize_google` and fix `require_api_key`
Josh Bickett committed
J
No `VERBOSE` needed here
Josh Bickett committed
January 15, 2024
J
add back clearing
Josh Bickett committed
J
Fix `Config` bug
Josh Bickett committed
J
fix validation bug
Josh Bickett committed
January 14, 2024
J
update `operate_type == "click"` condition
Josh Bickett committed
J
remove extra list dimension in `call_gemini_pro_vision`
Josh Bickett committed
J
Iterate `call_gpt_4_vision_preview_labeled`
Josh Bickett committed
January 13, 2024
J
Update to `operating_system.py`
Josh Bickett committed
J
Increase loop max
Josh Bickett committed
J
Add `config.verbose` and better `print`
Josh Bickett committed
J
Add missing `__init__.py`
Josh Bickett committed
J
Update some file names, add `get_user_prompt`
Josh Bickett committed
J
Add `ANSI_BLUE`
Josh Bickett committed
J
Add `operation.get("summmary")`
Josh Bickett committed