diff --git a/tts-server/app.py b/tts-server/app.py
index 27a6dc9..0e74061 100644
--- a/tts-server/app.py
+++ b/tts-server/app.py
@@ -50,8 +50,16 @@ log.info("Loading Kokoro (lang_code=%s) on %s", LANG_CODE, DEVICE)
 kokoro_pipeline = KPipeline(lang_code=LANG_CODE, device=DEVICE)
 log.info("Kokoro ready")
 
-log.info("Loading Sopro on %s", DEVICE)
-sopro_model = SoproTTS.from_pretrained("samuel-vitorino/sopro-v2-turbo", device=DEVICE)
+# Pinned instead of tracking the mutable `main` branch: upstream restructured
+# the repo on 2026-09-15 (dropped/renamed vocoder_streaming.safetensors and
+# speaker_encoder.safetensors), which crash-looped this container on every
+# boot since the unpinned load kept re-fetching the broken layout.
+SOPRO_REVISION = os.environ.get("TTS_SOPRO_REVISION", "ceeeb86b5fd805e662a5a750d6053ac4f990a45a")
+
+log.info("Loading Sopro on %s (revision=%s)", DEVICE, SOPRO_REVISION)
+sopro_model = SoproTTS.from_pretrained(
+    "samuel-vitorino/sopro-v2-turbo", device=DEVICE, revision=SOPRO_REVISION
+)
 log.info("Sopro ready")
 
 # References are just resampled/cropped tensors of the voice clip -- cheap to