WhisperSage 17+

Private Local AI Assistant

智慧 焦

Designed for iPad

    • $1.99

Screenshots

Description

Experience a personally tailored local AI assistant with LLMs such as Llama 3, Qwen2, managing conversations at any time.

Welcome to WhisperSage, your personal private local AI chat assistant.

Core Features:
- Conversation Management: Easily create, manage, and delete conversations.
- Privacy Protection: No need to connect to the internet, all conversations are stored locally, ensuring your information is secure and private, allowing you to use it with confidence.
- Model Management: Download models from Huggingface.co or use a remote model with ollama(https://ollama.com/, make sure ollama binds to 0.0.0.0 not localhost, and firewall is closed). Support both text model and multimodal model.
- Simple and Easy to Use: A user-friendly interface design makes managing conversations an easy and enjoyable task. Support speech input(speech to text, STT) with OpenAI Whisper local model.

IMPORTANT NOTE: Make sure you have enough memory to run the model, at least 1.5 times the size of the model (an additional 300MB for speech input and an additional 1.5GB for multimodal model). Considering user experience, the app is intend to run properly on iPhone 14 or newer devices, or devices with Apple Silicon M-series chips.

If you have any questions, please leave a message or send me an email.
Email:jzhihui521@gmail.com

What’s New

Version 1.2.2

- Update default model to Qwen2-1.5B-Instruct-Q4.
- Support download latest Qwen2 model: Qwen2-7B-Instruct-Q4.
- Fix potential bug.

App Privacy

The developer, 智慧 焦, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy.

Data Not Collected

The developer does not collect any data from this app.

Privacy practices may vary, for example, based on the features you use or your age. Learn More

Supports

  • Family Sharing

    Up to six family members can use this app with Family Sharing enabled.