Update SongGeneration MLX usage instructions

README.md

@@ -34,14 +34,15 @@ This is not a full-stack pure MLX audio pipeline yet: token generation runs with
 ## Quick Start
 
 ```bash
-hf download mlx-community/SongGeneration-v2-large-8bit --local-dir ./models/SongGeneration-v2-large-8bit
-
 git clone https://github.com/ailuntx/SongGeneration-MLX.git
 cd SongGeneration-MLX
 python -m venv .venv
 .venv/bin/pip install -e .
+.venv/bin/pip install -U huggingface_hub hf_transfer
+
+HF_HUB_ENABLE_HF_TRANSFER=1 .venv/bin/hf download mlx-community/SongGeneration-v2-large-8bit --local-dir ./models/SongGeneration-v2-large-8bit
 
-PYTHONPATH=. .venv/bin/python -m songgeneration_mlx.cli \
+.venv/bin/python -m songgeneration_mlx.cli \
   --model ./models/SongGeneration-v2-large-8bit \
   --lyrics '[verse] hello from mlx [chorus] sing it again' \
   --description 'Pop, female vocal, bright production, [Musicality-medium].' \
@@ -51,7 +52,29 @@ PYTHONPATH=. .venv/bin/python -m songgeneration_mlx.cli \
   --output tokens_2s.npz
 ```
 
-To decode tokens to FLAC, use the official decoder bridge documented in the GitHub README. The bridge requires the official SongGeneration runtime assets.
+## Decode Tokens
+
+The MLX runtime writes discrete song tokens. To create FLAC audio, use the official decoder bridge in [`ailuntx/SongGeneration-MLX`](https://github.com/ailuntx/SongGeneration-MLX). The bridge needs the official SongGeneration `runtime/` assets, but it does not need the original SongGeneration LM `model.pt` when `--mlx-model` is used.
+
+```bash
+python -m venv .venv-decoder
+.venv-decoder/bin/pip install -U pip
+.venv-decoder/bin/pip install \
+  -r third_party/SongGeneration/requirements.txt \
+  -r third_party/SongGeneration/requirements_nodeps.txt \
+  soundfile
+
+HF_HUB_ENABLE_HF_TRANSFER=1 .venv/bin/hf download tencent/SongGeneration \
+  --include "runtime/*" \
+  --local-dir ./third_party/SongGeneration
+
+PYTORCH_ENABLE_MPS_FALLBACK=1 SONGGEN_DEVICE=mps \
+.venv-decoder/bin/python scripts/decode_tokens_official.py \
+  --mlx-model ./models/SongGeneration-v2-large-8bit \
+  --tokens ./tokens_2s.npz \
+  --output ./output_2s.flac \
+  --device mps
+```
 
 ## Variants
 
@@ -65,6 +88,7 @@ To decode tokens to FLAC, use the official decoder bridge documented in the GitH
 | `SongGeneration-v2-large-bfloat16` | 9.5G | large quality baseline |
 | `SongGeneration-v2-large-8bit` | 5.0G | smaller large checkpoint |
 | `SongGeneration-v2-large-4bit` | 2.7G | smallest large checkpoint |
+| `SongGeneration-v2-fast-*` | pending | upstream fast weights were not publicly available when checked on 2026-05-31 |
 
 ## Layout