feat(mps): chunk the frame processor() call — bound the long/HD memory transient

fix(dedup): correct the cache-key rationale comment (Theory A proven)

feat(mps): dedup-cache machinery — hardened key, clear-per-Score, n_pass (gated)

feat(ui): timeline playhead follows video playback

feat(ui): in-card determinate progress bar via gr.Timer (Option T)

feat(ui): determinate per-clip encode progress bar (Option X)

refine(mps): blocked attention BLOCK=1024 + stock-fallback for edge cases (mask/causal/dropout/training/small-N)

feat(mps): blocked/streaming V-JEPA2 attention (Approach C) — structural memory bound, replaces per-layer empty_cache hook

fix(windowing): global z-score post-stitch (cross-window scale consistency) to cure WIN-1 seam

feat(quality): surface 'no speech track -> video-only' notice when no Word events

feat(mps): wire vm_stat memguard into the dedup encode loop (start gate + periodic abort)

perf(mps): HIGH watermark 0.6->0.45 (graceful OOM under the gate) + document B=1

feat(mps): per-layer empty_cache hook on V-JEPA2 encoder layers (TRIBE_MPS_EMPTY_EVERY)

feat(memguard): vm_stat system-memory governor for local MPS runs

feat(text): preview synthesized speech + remove profiler harness

feat: local MPS all-modes-working (audio + text); video deferred to HF

feat: local MPS port + ZeroGPU perf toolkit (dedup, backbone-reuse, pre-warm)

revert: disable torch.compile on V-JEPA2 (dynamo tracer bug broke video runs)

perf: torch.compile (static AOTInductor graph) on V-JEPA2 encode

fix(perf): batched encode B=4 + correct exca-descriptor fallback

perf+fix: batched V-JEPA2 encode (~3-8x) + writable hub cache for opt-in ASR/LLaMA

perf: bf16 V-JEPA2 encode via scoped monkeypatch (~2x, fp32 output)

perf: raise extractor batch sizes to exploit ZeroGPU VRAM (~48GB) + VRAM log

fix(zerogpu): force data.num_workers=0 so predict() works in the GPU worker

Integrate full app: 3-mode UI + windowed inference + ROI metrics + timeline

Scaffold TRIBE v2 Video Brain-Score Space