feat: add Modal deployment for FLUX.2 klein 9B image generation endpoint

untrack test image, ignore generated images

fix: improve weight directory validation to ensure critical shards exist before initialization and add test asset

docs: update README with comprehensive Aiko-chan project documentation and architecture details

feat: add FLUX.2 Klein image generation service and enable flash-attention for MioTTS

Update ui/css.py

Update app.py

Update app.py

Update app.py

Update app.py

Rename static/Aiko1.vrm to static/Aiko.vrm

Delete static/Aiko.vrm

Rename static/Miku.vrm to static/Aiko.vrm

Rename static/Aiko.vrm to static/Aiko1.vrm

Upload Miku.vrm

refactor: adjust mouth clamping, jaw rotation, and viseme mappings while removing redundant mesh double-sided rendering logic

fix: mitigate VRM chin mesh artifacts by clamping mouth weight, adjusting jaw rotation, forcing double-sided rendering, and refining viseme defaults

refactor: migrate weather provider to Open-Meteo, improve location extraction, and inject server time into persona context

fix: correct wttr.in endpoint usage and improve timezone resolution logic for diverse location inputs

feat: add support for immediate camera trigger and bypass LLM generation in think loop

feat: add camera auto-open signal handler to trigger vision modal from LLM output

style: fix comment formatting in app.py

feat: update capture_camera_image to trigger frontend camera modal via marker string

refactor: replace file-based vision uploads with base64 encoded data processing for improved compatibility

feat: add interactive webcam modal for photo capture and file uploads

chore: update Modal GPU requirement from T4 to A10G for MioTTS inference

refactor: remove unused imports from app.py

refactor: update ASR sample source to local static file

Add test voice

style: decrease max-height of aiko-info-card to 74vh

fix: reduce max-height of info card to 76vh to prevent overflow

style: increase max-height of aiko-info-card to 80vh

feat: add warning icon to privacy disclaimer in app UI

docs: update Aiko-chan description to include new search and vision capabilities

style: reduce max-height of aiko-info-card to 72vh

style: increase maximum width of aiko-info-card to 160vw

style: increase max width of aiko-info-card to 600px

fix: increase max-height of info card to 90vh to improve content visibility

feat: add MiniCPM-V 4.6 vision model details to soul persona documentation

feat: add camera button styling and update camera icon to frame emoji

feat: implement camera capture tool to enable vision capabilities via ffmpeg

feat: enhance service warmups with real-data validation and add VLM and SearXNG initialization steps

refactor: require essential environment variables by raising RuntimeErrors instead of using default values

refactor: relocate audio and perception logic to core module and remove legacy ui components

refactor: standardize ASR endpoint to /transcribe and update MiniCPM-V dependency and video decoding logic

feat: migrate ASR backend to Modal Nemotron endpoint with language support

feat: add MiniCPM-V 4.6 vision inference endpoint and update Nemotron-ASR to Python 3.11 with regex-based transcription parsing.

feat: transition to cache-aware streaming inference by cloning NeMo and running external inference script

feat: add Nemotron ASR backend for speech transcription and update persona skills to include Japanese teaching

feat: add crypto price tracking tool and update persona architecture documentation