Skip to content

The open-source AI voice studio. Clone, dictate, create.

fix(setup): install mlx-lm and mlx-audio in setup-python on Apple Silicon (#892)

Gregoire Moreelsf750596364d1
diff --git a/backend/requirements-mlx.txt b/backend/requirements-mlx.txt
index b1a82d5d..6699e588 100644
--- a/backend/requirements-mlx.txt
+++ b/backend/requirements-mlx.txt
@@ -16,7 +16,8 @@ miniaudio>=1.59
# mlx_audio.stt.load) works fine on transformers 4.57.x in practice.
#
# Install it via `pip install --no-deps mlx-audio==0.4.1` after this file
-# (see .github/workflows/release.yml). Most other mlx-audio runtime deps
+# (see .github/workflows/release.yml and the setup-python recipe in the
+# justfile). Most other mlx-audio runtime deps
# (huggingface_hub, librosa, mlx-lm, numba, numpy, protobuf, pyloudnorm,
# sounddevice, tqdm) are already in requirements.txt or pulled in by
# other engines.
diff --git a/backend/tests/test_mlx_smoke.py b/backend/tests/test_mlx_smoke.py
new file mode 100644
index 00000000..13c4b616
--- /dev/null
+++ b/backend/tests/test_mlx_smoke.py
@@ -0,0 +1,55 @@
+"""
+Smoke test for the MLX backend dependencies on Apple Silicon.
+
+Guards the `--no-deps` install of mlx-audio/mlx-lm done by `just setup-python`
+and release.yml: those packages skip their declared dependencies (transformers
+>=5.x conflict), so a missing transitive dep only surfaces at import time.
+This test fails fast if the MLX STT/TTS entry points the backend uses stop
+importing (e.g. the `miniaudio` regression from issue #505).
+
+Usage:
+ python -m pytest backend/tests/test_mlx_smoke.py -v
+"""
+
+import platform
+import sys
+
+import pytest
+
+pytestmark = pytest.mark.skipif(
+ not (sys.platform == "darwin" and platform.machine() == "arm64"),
+ reason="MLX packages are only installed on Apple Silicon macOS",
+)
+
+
+def test_mlx_core_runs():
+ """The MLX runtime itself works (Metal array op)."""
+ import mlx.core as mx
+
+ assert mx.array([1, 2]).sum().item() == 3
+
+
+def test_mlx_audio_tts_entry_point():
+ """`from mlx_audio.tts import load` — used by MLXBackend.load_model_async."""
+ from mlx_audio.tts import load
+
+ assert callable(load)
+
+
+def test_mlx_audio_stt_entry_point():
+ """`from mlx_audio.stt import load` — used by the Whisper MLX STT path.
+
+ Importing mlx_audio.stt also pulls in miniaudio, so this catches the
+ ModuleNotFoundError from issue #505 on fresh installs.
+ """
+ from mlx_audio.stt import load
+
+ assert callable(load)
+
+
+def test_mlx_lm_entry_points():
+ """`mlx_lm.load` / `mlx_lm.generate` — used by qwen_llm_backend."""
+ from mlx_lm import generate, load
+
+ assert callable(load)
+ assert callable(generate)
diff --git a/justfile b/justfile
index 4550d263..877f1519 100644
--- a/justfile
+++ b/justfile
@@ -72,6 +72,12 @@ setup-python:
if [ "$(uname -m)" = "arm64" ] && [ "$(uname)" = "Darwin" ]; then
echo "Detected Apple Silicon — installing MLX dependencies..."
{{ pip }} install -r {{ backend_dir }}/requirements-mlx.txt
+ # mlx-lm and mlx-audio declare transformers>=5.x, which conflicts with
+ # our transformers<=4.57.x cap, so install them --no-deps (their other
+ # runtime deps are covered by requirements.txt / requirements-mlx.txt —
+ # see the note in requirements-mlx.txt and .github/workflows/release.yml)
+ {{ pip }} install --no-deps mlx-lm==0.31.1
+ {{ pip }} install --no-deps mlx-audio==0.4.1
fi
{{ pip }} install git+https://github.com/QwenLM/Qwen3-TTS.git
{{ pip }} install pyinstaller ruff pytest pytest-asyncio -q