Spaces:
Building
Building
V3 Out
#5
by
AIencoder
- opened
๐ V3 Release: Vision, Voice, and a New UI!
V3 transforms the AI Coding Genius into a complete multi-modal assistant, all while remaining optimized for the Hugging Face Free CPU Tier.
โจ New Features:
- ๐ธ Vision Analyzer (OCR): Upload a screenshot of your error logs, and the AI will extract the text and fix it automatically.
- ๐๏ธ Auto-Transcribe: Record your logic or problem using the microphone, and Whisper will instantly convert it to text.
- ๐จ Visual Coder UI: A completely redesigned interface with a professional "Indigo Dark" theme and a clean sidebar layout.
๐ ๏ธ Technical Improvements:
- Unified Sidebar: Voice, Actions, and Vision tools are now stacked efficiently on the left.
- CPU Stability: Optimized threading to run Qwen 1.5B, Whisper, and Tesseract simultaneously without OOM crashes.
Try it out and let me know what you think!