Feat: Enable native parallelize=True flag for efficient batch persona generation

Fix: ensure robust person parsing

Fix: ensure sampled_characteristics is initialized to prevent UnboundLocalError during LLM retries

Feat: Implement dynamic validation and pipeline logic from IPYNB

Feat: Adapt tinytroupe_manager to use new persona_pipeline logic

Fix: Fallback to sequential persona generation with delay to prevent Google 429 Too Many Requests errors

Fix: revert google url to proper openai extension

Fix: remove trailing slash from google genai proxy base_url

Fix: brutally enforce base_url for Google GenAI in all OpenAI wrappers

Fix: allow token counting logic to intercept gemini model gracefully

Feat: Map default model to gemini-3-flash-preview

Feat: Map default model to gemini-3-flash-preview

Feat: Set Google Gemini GenerativeLanguage API endpoint as default

Fix: Strip empty `stop` parameter for Google Gemini fallback

Fix: Strip unsupported payload parameters for Google Gemini OpenAI-compatible proxy

Feat: Fallback to Google gemini-3-flash-preview on primary LLM proxy error

Fix: point model to alias-huge in nested config

Fix: point model to alias-huge

Fix: Fallback handling for empty JSON extraction in backend manager

Fix: Add missing dependency pypandoc for TinyTroupe ResultsExtractor

Refactor: Use TinyPersonFactory and ResultsExtractor for production LLM stability

Fix: combine system messages to satisfy strict LLM chatml parser

Fix: allow token counting logic to intercept alias models gracefully

Fix: remove 35s wait time throttle in nested config

Fix: remove 35s wait time throttle in config

Fix: enforce system message to be the first in openai_utils requests

Fix: revert model to alias-large in nested config

Fix: revert model to alias-large

Upload folder using huggingface_hub

Fix: Add local file persistence to JobRegistry to prevent 404s after space restart

Fix: revert MODEL to gpt-4o in nested config

Fix: revert MODEL to gpt-4o to resolve alias-large unavailability

Upload folder using huggingface_hub

Fix: remove alias-fast from nested config

Fix: remove alias-fast from root config

Fix: Replace deprecated alias-fast with alias-large in fallback cascade

Upload folder using huggingface_hub

Upload folder using huggingface_hub

Fix: map base_url explicitely to Helmholtz endpoint in OpenAI client constructor

Restart space to reload config

Fix: strictly map OPENAI_API_KEY to environment in config

Fix: Explicitly provide dummy token to OpenAI client constructor if environment vars not mapped early enough

Fix: map BLABLADOR_API_KEY to OPENAI_API_KEY in environment for TinyTroupe client validation

Fix: Append uuid to mock persona names to prevent TinyTroupe naming collision

Feat: Enable real LLM querying inside TinyTroupe simulation loops

Fix SyntaxWarning for invalid escape sequence in llm.py

Fix: bypass strict git clone check to prevent startup log errors

Fix ValueError for TOP_P parsing in tinytroupe init

Upload folder using huggingface_hub

Fix: Add missing dependencies (chevron, textdistance, llama-index) to requirements.txt