Allice
Basic information
Website: https://web.archive.org/web/20241123091858/https://github.com/myshell-ai/AIlice
Short description: AIlice is an open-source, fully autonomous AI assistant designed to performs complex tasks using LLMs. It supports multimodal and voice interactions and can dynamically expand its capabilities through modular configurations. [source]
Intended uses: What does the developer say it’s for? Research, coding, system management, literature reviews and “complex hybrid tasks” [source]
Date(s) deployed: June 22, 2024 [source]
Developer
Website: https://web.archive.org/web/20241123091858/https://github.com/myshell-ai/AIlice
Legal name: Myshell Limited [source] [source]
Entity type: Unknown
Country (location of developer or first author’s first affiliation): California, USA [source]
Safety policies: What safety and/or responsibility policies are in place? Unknown
System components
Backend model: What model(s) are used to power the system? Commercial LLMs like OpenAI’s GPT-4 and Anthropic’s Claude, as well as open-source models such as Qwen-2-72b-instruct, Mixtral, Meta LLaMA, and integrations with platforms like Ollama and LM Studio [source]
Publicly available model specification: Is there formal documentation on the system’s intended uses and how it is designed to behave in them? None
Reasoning, planning, and memory implementation: How does the system ‘think’? Decomposes tasks into subtasks managed by multiple collaborating agents using an Interactive Agents Call Tree architecture, while maintaining long-term memory through knowledge graphs and dynamic variable management [source]
Observation space: What is the system able to observe while ‘thinking’? Can observe local files, system processes, web content, and multimodal data (images, audio) with observations distributed across components that can run on different machines. The system maintains conversation context and can dynamically load new observation modules, allowing it to expand its observation capabilities over time [source]
Action space/tools: What direct actions can the system take? AIlice can execute code, manage files and system processes, analyze data, generate visualizations, interact with web services, process multimodal content (images/audio), and dynamically expand its capabilities by building new interaction modules [source]
User interface: How do users interact with the system? Web-based dialogue interface accessible through a browser, allowing users to interact using text or voice commands to issue tasks and receive responses [source]
Development cost and compute: What is known about the development costs? Unknown
Guardrails and oversight
Accessibility of components:
- Weights: Are model parameters available? Unknown
- Data: Is data available? Unknown
- Code: Is code available? Available [source]
- Scaffolding: Is system scaffolding available? Available [source]
- Documentation: Is documentation available? Available [source]
Controls and guardrails: What notable methods are used to protect against harmful actions? Allows for user-interrupt mechanisms, running tasks within secure Docker containers, and warns user to engage in active monitoring [source]
Customer and usage restrictions: Are there know-your-customer measures or other restrictions on customers? None
Monitoring and shutdown procedures: Are there any notable methods or protocols that allow for the system to be shut down if it is observed to behave harmfully? Enables users to manually interrupt its operations through a web-based interface [source]
Evaluation
Notable benchmark evaluations: None
Bespoke testing: Demos [source]
Safety: Have safety evaluations been conducted by the developers? What were the results? None
Publicly reported external red-teaming or comparable auditing:
- Personnel: Who were the red-teamers/auditors? None
- Scope, scale, access, and methods: What access did red-teamers/auditors have and what actions did they take? None
- Findings: What did the red-teamers/auditors conclude? None
Ecosystem information
Interoperability with other systems: What tools or integrations are available? Integrates with external APIs, Docker containers, inference services like OpenRouter, and supports dynamic module loading through its configuration system [source]
Usage statistics and patterns: Are there any notable observations about usage? Github repo has 869 stars and 134 forks [source]
Additional notes
None