Spaces:
Sleeping
Sleeping
Commit History
-
Force HF Spaces rebuild: Add terminal status to init log
lowvoltagenation +1: ClaudeSmall change to trigger HuggingFace Spaces rebuild since the previous changes for expanded terminal by default may not have been detected. This commit adds a status message to confirm terminal is expanded by default. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>d5802dd -
Make terminal expanded by default for better website embed experience
lowvoltagenation +1: Claude🎯 Changes for website embed: - Terminal now expanded/visible by default instead of collapsed - Checkbox starts checked (Show Real-time Logs = true) - Updated welcome message to be more engaging since it's always visible - Better first impression for website visitors 🌟 Benefits: - Immediate visual impact showing Maya's sophisticated backend - Visitors can instantly see system activity without interaction - Perfect for demonstrating AI capabilities in embedded format - More engaging and impressive for technical audiences The terminal gives immediate insight into Maya's AI operations, making the embed much more compelling for website visitors. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>dfe9a8b -
Fix terminal issues: persistent logs + force HF Spaces rebuild
lowvoltagenation +1: Claude🔧 Fixes: 1. **Log Preservation**: Replace queue with persistent list + threading.Lock - Logs now persist throughout session instead of being consumed - Thread-safe access with proper locking - Stores last 200 entries, displays last 100 2. **HF Spaces Deployment**: Force fresh rebuild - HF Spaces was still using old gr.Code version despite gr.Textbox fix - Small change to trigger complete rebuild and cache refresh - Should resolve the 'Language bash not supported' error 3. **Improved Log Storage**: - Uses list instead of queue for persistent storage - Thread-safe with proper locking mechanism - Better memory management with rolling window The terminal will now maintain log history throughout the session and HF Spaces should properly deploy with gr.Textbox component. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>32f9ea8 -
Fix terminal component: replace gr.Code with gr.Textbox
lowvoltagenation +1: Claude- Replaced gr.Code(language='bash') with gr.Textbox to fix unsupported language error - Updated CSS to properly style textarea element for terminal appearance - Added show_copy_button=True for better usability - Maintained black terminal background with green text styling - Fixed Gradio 5.0 compatibility issue The terminal now uses a standard textbox with terminal styling instead of the code component that doesn't support bash syntax highlighting. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>c3e3fd8 -
Add comprehensive real-time terminal for Maya backend demo
lowvoltagenation +1: Claude✨ Features: - Real-time terminal showing all system activity and logs - Collapsed by default for clean demo experience - Auto-refresh every 2 seconds when visible - Color-coded log levels (INFO=green, ERROR=red, etc.) - Captures ALL backend activity: model loading, chat generation, RAG retrieval - Enhanced logging throughout the system with emojis for clarity - Terminal-style CSS with black background and green text - Manual refresh button + auto-refresh capability 🎯 Demo Benefits: - Users can see Maya's 'thinking' process in real-time - Shows model loading progress and authentication steps - Displays RAG memory retrieval and storage - Reveals response generation parameters and timing - Perfect for technical demonstrations and debugging The terminal captures everything from model loading to response generation, giving full visibility into Maya's backend operations. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>028cec7 -
Add missing sentencepiece dependency and improve tokenizer loading
lowvoltagenation +1: Claude- Added sentencepiece>=0.1.99 to requirements.txt (required for Mistral tokenizer) - Added protobuf and tiktoken dependencies for better tokenizer support - Improved tokenizer loading with fast/slow fallback logic - Maya LoRA should now load properly with working tokenizer The model was loading (checkpoint shards 100%) but tokenizer was failing. This should resolve the sentencepiece tokenizer error. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>1c69287 -
Restore Maya LoRA for proper gated model access with Pro account
lowvoltagenation +1: Claude- Revert Maya LoRA to inference_api type with requires_auth: True - Use original Mistral v0.3 base model as intended - User has HuggingFace Pro account and should be able to access gated models - Need to configure HF_TOKEN in Spaces secrets for authentication Next step: Add HF_TOKEN secret in HuggingFace Spaces settings. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>45bd7a8 -
Add LoRA fallback system for Maya model compatibility
lowvoltagenation +1: Claude- Switch Maya LoRA from inference_api to local LoRA loading - Add fallback base model system (v0.1 first, v0.3 if needed) - Add _load_as_inference_api() fallback method if LoRA fails - This should handle the gated v0.3 issue by trying non-gated v0.1 first - Provides multiple fallback paths for Maya LoRA loading The system will try: 1. Load LoRA with non-gated Mistral v0.1 base 2. Fallback to gated Mistral v0.3 base if needed 3. Fallback to inference API if LoRA loading fails entirely 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>cbc3ed8 -
Fix Maya LoRA to use non-gated Mistral v0.1 base model
lowvoltagenation +1: Claude- Updated Maya LoRA base_model reference from v0.3 to v0.1 - Mistral v0.3 became gated and requires approval - Mistral v0.1 is non-gated and should work without authentication - Maya LoRA should now load properly via inference API 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>6ce1953 -
Restore Maya LoRA model: Use inference API to avoid base model download issues
lowvoltagenation358886d -
Fix gated model issue: Use Mistral v0.1 with Maya personality via system prompt
lowvoltagenationcb29bc9 -
Fix HF Spaces auth: Use inference API for Maya model to avoid gated base model download
lowvoltagenationfca6913 -
Add new user interaction memory for Maya: greeting response and programming assistance
lowvoltagenation4e11a03 -
Fix Maya LoRA responses: simplify prompts, add proper formatting, improve cleaning
lowvoltagenation6895a87 -
Add support for LoRA model loading in ModelInterface
lowvoltagenation- Updated requirements.txt to include 'peft' library. - Enhanced ModelInterface to load LoRA adapters with base models, including error handling and tokenizer setup. - Integrated logging for model loading processes to improve feedback during operations.2043924 -
Update model dropdown to show only Maya LoRA, Mistral v0.3, and Kimi K2
lowvoltagenationed31be4
-
Refactor load_model method in MayaGradioApp to simplify return values and enhance model loading feedback. Update UI elements for model selection and settings, improving layout and user interaction.
lowvoltagenation5457f70 -
Refactor chat layout in MayaGradioApp by adding new CSS classes for chat and settings columns, and enhancing input row styling for improved user interaction.
lowvoltagenationff8a780 -
Enhance header styling in MayaGradioApp by adjusting text alignment and padding, and adding new subtitle and tagline elements for improved branding and presentation.
lowvoltagenationa43ab37 -
Update MayaCharacter prompts to replace references to 'Blake' with 'User' and remove explicit content restrictions for improved interaction guidelines.
lowvoltagenation1a39c1f -
Refactor chat history structure in MayaGradioApp to use dictionaries for messages and enhance CSS styling for the header.
lowvoltagenationc52779e
-
adding maya-gradio files
ba20783