Update src/llm/local_model_client.py

Update src/llm/local_model_client.py

Update requirements.txt

Update src/llm/local_model_client.py

Update src/llm/local_model_client.py

Add HuggingFace token support for gated Llama model access

Add peft package for Be.FM PEFT adapter loading

Make opinion parser more flexible for Be.FM and other LLMs

Fix Be.FM to support CUDA and CPU in addition to MPS

Add PyTorch and transformers for Be.FM local model support

Fix persona IDs to match database and collapse diagnostics

Fix population network distribution and add opinion-based node coloring

Add persona_weights support to PopulationNetwork

Fix Edit Personas income_level error and add missing VariantGenerator class

Fix multiple bugs: text_area height, Pydantic schema, double-click, sidebar

Change model to Claude 3 Haiku for cost efficiency

Fix model 404 error: Use claude-3-5-sonnet-20240620 instead of 20241022

FINAL FIX: Remove st.rerun() to eliminate white screen flashing

Add HuggingFace Spaces configuration to README.md

Fix white screen flashing by removing value parameter and st.rerun()

Fix session state error on main page example questions

Add custom page labels via pages.toml configuration

Fix Streamlit session state errors and UI improvements

Add automated deployment fix script for HF Spaces

Add LLM initialization test script for debugging

Add deployment status and verification checklist for HF Spaces fix

Add comprehensive error display and diagnostics for debugging

Trigger HF Spaces rebuild with updated requirements

Fix HF Spaces initialization and add comprehensive diagnostics

Improve LLM client error handling and API key detection

Use explicit httpx client to avoid proxies parameter error

Add version compatibility handling for Anthropic client proxies error

Add manual deployment fix script for HF Spaces

Fix IndentationError and proxies parameter issue in AnthropicClient

Fix NameError for llm_client variable

Fix Anthropic client initialization error

Fix session state KeyError in web_app.py

Fix module import errors in web_app.py

Add Hugging Face Spaces deployment script and gitignore

Fix example questions clearing bug - use widget key directly

Fix Population Analysis bugs: question persistence and position detection

Optimize Be.FM MPS performance - 17x speedup (5min → 18sec)

Add Hugging Face Spaces deployment configuration

Fix Population Analysis UX and Be.FM concurrency issues

Add Edit Personas page and complete model selector integration

Fix watermark ratio error and default to MPS GPU

Apply comprehensive MPS stability improvements for M4 Mac

Merge PEFT weights to enable stable MPS GPU inference

Use CPU by default for Be.FM to avoid MPS segmentation faults

Add thread-safe inference to prevent MPS segmentation faults