Update bf16 naming and collection links

README.md

@@ -12,9 +12,9 @@ tags:
 - audiolm
 ---
 
-Part of the SongGeneration MLX conversion set.
+Part of the SongGeneration MLX conversion set. Collection: https://huggingface.co/collections/mlx-community/songgeneration-v2-mlx-6a1bf9342dd0806419737229
 
-# SongGeneration-v2-medium-bfloat16
+# SongGeneration-v2-medium-bf16
 
 Apple MLX weights for the autoregressive `audiolm` token generator from Tencent SongGeneration v2-medium.
 
@@ -25,7 +25,7 @@ This is not a full-stack pure MLX audio pipeline yet: token generation runs with
 | | |
 |---|---|
 | **Variant** | `v2-medium` |
-| **Precision** | `bfloat16` |
+| **Precision** | `bf16` (`bfloat16`) |
 | **Converted component** | SongGeneration `audiolm` token generator |
 | **Runtime** | [`ailuntx/SongGeneration-MLX`](https://github.com/ailuntx/SongGeneration-MLX) |
 | **Official model** | [`tencent/SongGeneration`](https://huggingface.co/tencent/SongGeneration) |
@@ -40,10 +40,10 @@ 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-medium-bfloat16 --local-dir ./models/SongGeneration-v2-medium-bfloat16
+HF_HUB_ENABLE_HF_TRANSFER=1 .venv/bin/hf download mlx-community/SongGeneration-v2-medium-bf16 --local-dir ./models/SongGeneration-v2-medium-bf16
 
 .venv/bin/python -m songgeneration_mlx.cli \
-  --model ./models/SongGeneration-v2-medium-bfloat16 \
+  --model ./models/SongGeneration-v2-medium-bf16 \
   --lyrics '[verse] hello from mlx [chorus] sing it again' \
   --description 'Pop, female vocal, bright production, [Musicality-medium].' \
   --duration 2 \
@@ -70,7 +70,7 @@ HF_HUB_ENABLE_HF_TRANSFER=1 .venv/bin/hf download tencent/SongGeneration \
 
 PYTORCH_ENABLE_MPS_FALLBACK=1 SONGGEN_DEVICE=mps \
 .venv-decoder/bin/python scripts/decode_tokens_official.py \
-  --mlx-model ./models/SongGeneration-v2-medium-bfloat16 \
+  --mlx-model ./models/SongGeneration-v2-medium-bf16 \
   --tokens ./tokens_2s.npz \
   --output ./output_2s.flac \
   --device mps
@@ -81,11 +81,11 @@ PYTORCH_ENABLE_MPS_FALLBACK=1 SONGGEN_DEVICE=mps \
 | Variant | Disk | Notes |
 |---|---:|---|
 | `SongGeneration-v2-medium-fp32` | 10G | high-precision medium baseline |
-| `SongGeneration-v2-medium-bfloat16` | 5.2G | recommended medium quality baseline |
+| `SongGeneration-v2-medium-bf16` | 5.2G | recommended medium bf16 quality baseline |
 | `SongGeneration-v2-medium-8bit` | 2.8G | smaller medium checkpoint |
 | `SongGeneration-v2-medium-4bit` | 1.5G | smallest medium checkpoint |
 | `SongGeneration-v2-large-fp32` | 19G | high-precision large baseline |
-| `SongGeneration-v2-large-bfloat16` | 9.5G | large quality baseline |
+| `SongGeneration-v2-large-bf16` | 9.5G | large bf16 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 |
@@ -93,7 +93,7 @@ PYTORCH_ENABLE_MPS_FALLBACK=1 SONGGEN_DEVICE=mps \
 ## Layout
 
 ```text
-SongGeneration-v2-medium-bfloat16/
+SongGeneration-v2-medium-bf16/
 |-- model-00001-of-000xx.safetensors
 |-- model.safetensors.index.json
 |-- config.json
@@ -105,7 +105,7 @@ SongGeneration-v2-medium-bfloat16/
 
 ## Validation
 
-Local Apple Silicon validation was run on the medium bfloat16 path:
+Local Apple Silicon validation was run on the medium bf16 path:
 
 | Test | Result |
 |---|---|