Instructions to use mlx-community/SongGeneration-v2-large-fp32 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/SongGeneration-v2-large-fp32 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir SongGeneration-v2-large-fp32 mlx-community/SongGeneration-v2-large-fp32
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
Update bf16 naming and collection links
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ tags:
|
|
| 12 |
- audiolm
|
| 13 |
---
|
| 14 |
|
| 15 |
-
Part of the SongGeneration MLX conversion set.
|
| 16 |
|
| 17 |
# SongGeneration-v2-large-fp32
|
| 18 |
|
|
@@ -81,11 +81,11 @@ PYTORCH_ENABLE_MPS_FALLBACK=1 SONGGEN_DEVICE=mps \
|
|
| 81 |
| Variant | Disk | Notes |
|
| 82 |
|---|---:|---|
|
| 83 |
| `SongGeneration-v2-medium-fp32` | 10G | high-precision medium baseline |
|
| 84 |
-
| `SongGeneration-v2-medium-
|
| 85 |
| `SongGeneration-v2-medium-8bit` | 2.8G | smaller medium checkpoint |
|
| 86 |
| `SongGeneration-v2-medium-4bit` | 1.5G | smallest medium checkpoint |
|
| 87 |
| `SongGeneration-v2-large-fp32` | 19G | high-precision large baseline |
|
| 88 |
-
| `SongGeneration-v2-large-
|
| 89 |
| `SongGeneration-v2-large-8bit` | 5.0G | smaller large checkpoint |
|
| 90 |
| `SongGeneration-v2-large-4bit` | 2.7G | smallest large checkpoint |
|
| 91 |
| `SongGeneration-v2-fast-*` | pending | upstream fast weights were not publicly available when checked on 2026-05-31 |
|
|
@@ -105,7 +105,7 @@ SongGeneration-v2-large-fp32/
|
|
| 105 |
|
| 106 |
## Validation
|
| 107 |
|
| 108 |
-
Local Apple Silicon validation was run on the medium
|
| 109 |
|
| 110 |
| Test | Result |
|
| 111 |
|---|---|
|
|
|
|
| 12 |
- audiolm
|
| 13 |
---
|
| 14 |
|
| 15 |
+
Part of the SongGeneration MLX conversion set. Collection: https://huggingface.co/collections/mlx-community/songgeneration-v2-mlx-6a1bf9342dd0806419737229
|
| 16 |
|
| 17 |
# SongGeneration-v2-large-fp32
|
| 18 |
|
|
|
|
| 81 |
| Variant | Disk | Notes |
|
| 82 |
|---|---:|---|
|
| 83 |
| `SongGeneration-v2-medium-fp32` | 10G | high-precision medium baseline |
|
| 84 |
+
| `SongGeneration-v2-medium-bf16` | 5.2G | recommended medium bf16 quality baseline |
|
| 85 |
| `SongGeneration-v2-medium-8bit` | 2.8G | smaller medium checkpoint |
|
| 86 |
| `SongGeneration-v2-medium-4bit` | 1.5G | smallest medium checkpoint |
|
| 87 |
| `SongGeneration-v2-large-fp32` | 19G | high-precision large baseline |
|
| 88 |
+
| `SongGeneration-v2-large-bf16` | 9.5G | large bf16 quality baseline |
|
| 89 |
| `SongGeneration-v2-large-8bit` | 5.0G | smaller large checkpoint |
|
| 90 |
| `SongGeneration-v2-large-4bit` | 2.7G | smallest large checkpoint |
|
| 91 |
| `SongGeneration-v2-fast-*` | pending | upstream fast weights were not publicly available when checked on 2026-05-31 |
|
|
|
|
| 105 |
|
| 106 |
## Validation
|
| 107 |
|
| 108 |
+
Local Apple Silicon validation was run on the medium bf16 path:
|
| 109 |
|
| 110 |
| Test | Result |
|
| 111 |
|---|---|
|