Fix: pre-download whisperx model during Docker build to avoid 429 rate limits at runtime

Dockerfile

@@ -20,6 +20,10 @@ RUN pip install --no-cache-dir \
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Pre-download whisperx model (public repo, no token needed)
+# This avoids 429 rate limits when uvx whisperx runs at inference time
+RUN python3 -c "from huggingface_hub import snapshot_download; snapshot_download('Systran/faster-whisper-large-v3', cache_dir='/tmp/hf_hub_cache')"
+
 COPY . .
 
 RUN useradd -m -u 1000 appuser && chown -R appuser:appuser /app
@@ -33,6 +37,7 @@ USER appuser
 
 ENV HF_HUB_CACHE=/tmp/hf_hub_cache
 ENV WHISPER_CACHE_DIR=/tmp/whisper_cache
+ENV HF_HUB_DISABLE_TELEMETRY=1
 
 EXPOSE 7860
 CMD ["python", "app.py"]

app.py

@@ -40,10 +40,13 @@ HF_TOKEN = os.environ.get("HF_TOKEN", "")
 CKPT = os.environ.get("TRIBE_CKPT", "facebook/tribev2")
 MAX_VIDEO_DURATION = int(os.environ.get("MAX_VIDEO_DURATION", "120"))  # seconds
 CACHE_DIR = os.environ.get("SGP_CACHE_DIR", "/tmp/sgp_atlas")
-HF_HUB_CACHE = os.environ.get("HF_HUB_CACHE", "/tmp/hf_hub_cache")
-os.environ["HF_HUB_CACHE"] = HF_HUB_CACHE
-os.environ["WHISPER_CACHE_DIR"] = os.environ.get("WHISPER_CACHE_DIR", "/tmp/whisper_cache")
-os.makedirs(HF_HUB_CACHE, exist_ok=True)
+
+# Ensure HF cache env vars are set for all subprocesses (including uvx whisperx)
+os.environ.setdefault("HF_HUB_CACHE", "/tmp/hf_hub_cache")
+os.environ.setdefault("WHISPER_CACHE_DIR", "/tmp/whisper_cache")
+os.environ.setdefault("HF_HUB_DISABLE_TELEMETRY", "1")
+os.makedirs(os.environ["HF_HUB_CACHE"], exist_ok=True)
+os.makedirs(os.environ["WHISPER_CACHE_DIR"], exist_ok=True)
 
 # ─── Result storage (in-memory for now; extend to file/DB for persistence) ───
 _stimulus_results = {}
@@ -66,27 +69,12 @@ def _load_model():
         if HF_TOKEN:
             os.environ["HUGGING_FACE_HUB_TOKEN"] = HF_TOKEN
             os.environ["HF_TOKEN"] = HF_TOKEN
-            os.environ["HF_HUB_CACHE"] = "/tmp/hf_hub_cache"
-            os.makedirs("/tmp/hf_hub_cache", exist_ok=True)
             try:
                 from huggingface_hub import login
                 login(token=HF_TOKEN, add_to_git_credential=False)
                 print(f"[SGP-Tribe3] HF login OK", flush=True)
             except Exception as e:
                 print(f"[SGP-Tribe3] HF login warning: {e}", flush=True)
-
-            # Pre-download whisperx model to avoid rate limits in uvx subprocess
-            try:
-                from huggingface_hub import snapshot_download
-                print("[SGP-Tribe3] Pre-downloading faster-whisper-large-v3...", flush=True)
-                snapshot_download(
-                    repo_id="Systran/faster-whisper-large-v3",
-                    cache_dir="/tmp/hf_hub_cache",
-                    token=HF_TOKEN
-                )
-                print("[SGP-Tribe3] Whisper model cached successfully", flush=True)
-            except Exception as e:
-                print(f"[SGP-Tribe3] Whisper model download warning: {e}", flush=True)
         else:
             print("[SGP-Tribe3] WARNING: No HF_TOKEN set — LLaMA encoder may fail", flush=True)