Releases: trycua/cua
Releases · trycua/cua
cua-computer-server v0.1.20
cua-computer-server v0.1.20
Replaced pyautogui.type_text with pynput.KeyboardController, fixing the bug where multi-line text would just type the letter "v" on Linux
cua-agent v0.4.5
cua-agent v0.4.5
Dependencies
- cua-computer: latest
- cua-som: latest
Installation Options
Basic installation with Anthropic
pip install cua-agent[anthropic]==0.4.5With SOM (recommended)
pip install cua-agent[som]==0.4.5All features
pip install cua-agent[all]==0.4.5cua-agent v0.4.4
cua-agent v0.4.4
Bugfixes caused by mouse_move having the wrong name
cua-agent v0.4.3
cua-agent v0.4.3
Bugfixes caused by mouse_move having the wrong name
cua-agent v0.4.2
cua-agent v0.4.2
What's Changed
- [Agent] Fix Anthropic unexpected tool_use_id by @ddupont808 in #325
Full Changelog: core-v0.1.7...agent-v0.4.2
cua-core v0.1.6
cua-core v0.1.6
Fixed bug where Posthog has no attribute 'identify' would be logged upon import
cua-computer v0.4.0
cua-computer v0.4.0
Bumped cua-computer to use latest cua-core
cua-agent v0.4.1
cua-agent v0.4.1
Updated to use latest cua-core
cua-agent v0.4.0
cua-agent v0.4.0
This update refactored the Agent SDK to make it easier to implement new features and support the release of new agent models/loops.
Changelog:
- Reworked agent loop, now all agent providers share a loop (Generate, Execute, Repeat), with the only difference between loops being the implementation of the Generate function
- Replaced LLM clients with LiteLLM, now all agent providers support any provider supported by LiteLLM
- Added 2 custom LiteLLM providers for local model inference on CUDA and MLX devices:
huggingface-local/,mlx/ - Reworked callback system to have hooks at every step of the lifecycle
- Converted logging, trajectory saving, image retention into callbacks
- Added new callbacks - PII Anonymization (still a W.I.P) & budget management
- Anthropic providers - Added support for explicit prompt caching
- OpenAI providers - Added support for zero data retention
- Added Agent CLI for quick testing:
python -m agent.cli <model name>
Breaking Changes
- Initialization:
ComputerAgent(v0.4.x) usesmodelas a string (e.g. "anthropic/claude-3-5-sonnet-20241022") instead ofLLMandAgentLoopobjects.toolsis a list (can include multiple computers and decorated functions).callbacksare now first-class for extensibility (image retention, budget, trajectory, logging, etc).
- No explicit
loopparameter:- Loop is inferred from the
modelstring (e.g.anthropic/,openai/,omniparser+,ui-tars).
- Loop is inferred from the
- No explicit
computerparameter:- Computers are added to
toolslist.
- Computers are added to
Install
# Before merge:
pip install --pre "cua-agent[all]==0.4.0b4"
# After merge:
pip install "cua-agent[all]"
# or install specific providers
pip install "cua-agent[openai]" # OpenAI computer-use-preview support
pip install "cua-agent[anthropic]" # Anthropic Claude support
pip install "cua-agent[omni]" # Omniparser + any LLM support
pip install "cua-agent[uitars]" # UI-TARS
pip install "cua-agent[uitars-mlx]" # UI-TARS + MLX support
pip install "cua-agent[uitars-hf]" # UI-TARS + Huggingface support
pip install "cua-agent[ui]" # Gradio UI supportSupported Models
Anthropic Claude (Computer Use API)
model="anthropic/claude-3-5-sonnet-20241022"
model="anthropic/claude-3-5-sonnet-20240620"
model="anthropic/claude-opus-4-20250514"
model="anthropic/claude-sonnet-4-20250514"OpenAI Computer Use Preview
model="openai/computer-use-preview"UI-TARS (Local or Huggingface Inference)
model="huggingface-local/ByteDance-Seed/UI-TARS-1.5-7B"
model="ollama_chat/0000/ui-tars-1.5-7b"Omniparser + Any LLM
model="omniparser+ollama_chat/mistral-small3.2"
model="omniparser+vertex_ai/gemini-pro"
model="omniparser+anthropic/claude-3-5-sonnet-20241022"
model="omniparser+openai/gpt-4o"cua-computer v0.3.7
cua-computer v0.3.7
Computer control library for the Computer Universal Automation (CUA) project
Dependencies
- pylume: latest
Installation
pip install cua-computer==0.3.7