chore: generate data from llm

chore: install llama-cpp from prebuilt Space wheel

feat: add local actor model backends

finetune: add GGUF publish prep workflow

finetune: add GGUF eval harness with chatml prompt

finetune: add merge and GGUF conversion tooling

docs: add published fine-tuning artifact links

finetune: add Hugging Face publish tooling

finetune: add Actor SFT v1 audit and eval hardening

finetune: add Actor SFT v1 hardening path

finetune: add Actor LoRA eval pipeline

finetune: add Modal actor LoRA training pipeline

finetune: add optional actor SFT seed ingestion

finetune: add actor SFT v0 dataset pipeline

docs: add table of contents in readme

docs: add rendered architecture diagrams

docs: update readme with architecture diagram and other details

fix: escape actor prompt tool examples

fix: cleanup checkboxes and change local openbmb behavior for full run

fix: disable ssr mode

feat: add agent trace export

feat: add edge tts voice mode

feat: add browser tts and conversation pacing

feat: add theatrical agent tools

feat: add agentic actor state

feat: add dynamic story pacing

feat: add optional llm director

fix: support python 3.10 space build

chore: prepare zerogpu runtime

feat: add trace export

feat: add hugging face inference backend

feat: add optional llm director

feat: add director policy

ui-changes (#5)

fix: harden OpenBMB backend runtime

feat: add optional OpenBMB actor backend

refactor: add model backend foundation

fix: tighten puppet stage layout

feat: polish puppet stage UI

feat: add audience actions

feat: add deterministic skit playback

feat: add puppet theater session models

refactor: pin python version in gradio space and fix pydantic's version

refactor: replace chatbot template with puppet theater shell

chore: add uv setup

docs: add agent instructions

initial commit

Duplicate from gradio-templates/chatbot