Track playback and key toggle state

Improve playback loop smoothness

Graceful fallback when motion_correction missing

Graceful fallback when motion_correction missing

Fix LLM2Vec PEFT adapter stacking

ZeroGPU-safe startup device + llm2vec tensor warning cleanup

ZeroGPU-safe startup device + llm2vec tensor warning cleanup

chore: suppress websocket probe handshake tracebacks

fix: prefer local text encoder API in api mode when available

fix: disable gradio hot reload for CUDA-safe startup

fix: fallback to local text encoder API when remote URL times out

fix: HTTP readiness gate in _ensure_text_encoder_server

fix: lazy TextEncoderAPI client with retry + HTTP readiness gate

Set Space SDK to gradio for zero-a10g compatibility

Add Space metadata and deployment updates

Cloud Run encoder wiring + startup resilience

Rebrand top-left UI label to Movimento

Fix text encoder payload parsing and add remote encoder switch

Switch to LLM2Vec 3.1 pair to fix Space 401 on gated Llama 3.0

Sync text encoder 3.0 fix and smoke checks

Fix HF auth propagation for text encoder loads

Fix: explicit HF_TOKEN passing to text encoder subprocess

Switch LLM2Vec to Meta-Llama-3.1-8B-Instruct (accessible with current token)

Improve error extraction: collect all error strings from text encoder response

Revert early spaces import in text encoder server (fix gradio_auto_wrap crash)

Spaces: import spaces early in text-encoder server to avoid CUDA init warning thread crash

Use mirrored LLM2Vec base model repo and env overrides

Debug: print full traceback on text encoder init failure to expose real error

Improve text encoder error messages: better diagnostics for HF token issues

Improve text encoder API error detection: validate .npy paths and detect error messages

Fix generate notification: reuse persistent notification and properly transition to error state

Space: keep existing embedding pipeline and start local text-encoder API server on boot

UI: remove debug markdown, orphaned button handler, lower default diffusion steps to 50

UI: remove example 09 pinning from dropdown and remove direct Load Qwen Example button

Space: invoke @spaces.GPU function during startup to satisfy ZeroGPU policy

UI: pin 09_qwen_agentic_actions at top of Examples dropdown and default-select it

Space: restore stable native-direct startup; keep explicit Qwen 09 controls

Space: redirect to native proxy without iframe recursion; keep Qwen 09 direct load

Space: restore minimal @spaces.GPU hook for ZeroGPU with native-direct UI

Space: run native Kimodo UI directly on PORT to avoid iframe recursion

Space: auto-retry native proxy load and add direct Qwen example button

Space: force native UI proxy and remove fallback iframe switching

Default HF runtime to CPU and honor KIMODO_DEVICE

Use lazy TextEncoderAPI client in API-only mode (no startup fallback)

Force TEXT_ENCODER_MODE=api on Space to avoid gated local fallback

Log imported kimodo module path in Space boot

Add skeleton loader entry-exit diagnostics and robust fallback

Add packaged demo and skeleton assets required for native startup

Add entry/exit debug logs for example dropdown lifecycle

Sync latest Qwen example dropdown fixes and add 09 example assets