Downloads · 30 days
5
15% of all-time downloads
soda-research/discrete-audio-isoflop-3e20-231M-d640-L7-B256-14d478
discrete-audio-isoflop-3e20-231M-d640-L7-B256-14d478 is a machine learning model from soda-research. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
A suite of discrete audio models trained for our IsoFLOP study as part of SODA, which is a unified next-token prediction on interleaved semantic, acoustic, and text tokens.
Downloads · 30 days
5
15% of all-time downloads
All-time downloads
34
Public
Parameters
231M
944 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors924 MB · 98%
From the Hugging Face model README
A suite of discrete audio models trained for our IsoFLOP study as part of SODA, which is a unified next-token prediction on interleaved semantic, acoustic, and text tokens.
🥤 Project Page: https://soda-audio.github.io
For full usage instructions (e.g., inference code), and more information, please refer to the SODA-4B-base model card.
The details for this particular model is as follows:
compute_budget: 3e20param_count (non-embedding): 231Mhidden_dim: 640num_layers: 7batch_size: 256training_step: 271914hash_key: 14d478📈 WandB: https://wandb.ai/potsawee/marin/groups/IsoFlop/workspace