qwen2_5_omni_all_1015_cleaned

boyuzhuGPT/qwen2_5_omni_all_1015_cleaned

qwen2_5_omni_all_1015_cleaned is a text-to-audio model from boyuzhuGPT. Use it for the text-to-audio task on the model card, and read the license before you ship it in a product. It is set up for transformers.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

56% of all-time downloads

All-time downloads

Public

Parameters

8.9B

482 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors17.9 GB · 100%

At a glance

Task: Text-to-Audio
Library: transformers
Model type: qwen2_5_omni
Access: Public
Created: Oct 15, 2025
Updated: Oct 15, 2025
SHA: 4d4e577e

Model card

From the Hugging Face model README

qwen2_5_omni_all_1015_cleaned

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6452
Token Acc: 0.7996

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0001
train_batch_size: 2
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 4
total_train_batch_size: 64
total_eval_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.05
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Token Acc
0.8822	0.1139	50	0.9159	0.7301
0.9312	0.2278	100	0.9529	0.7280
0.9248	0.3417	150	0.9446	0.7316
0.9002	0.4556	200	0.9342	0.7329
0.8913	0.5695	250	0.9067	0.7399
0.8639	0.6834	300	0.8917	0.7419
0.8537	0.7973	350	0.8809	0.7453
0.8474	0.9112	400	0.8541	0.7517
0.7713	1.0251	450	0.8413	0.7543
0.7516	1.1390	500	0.8243	0.7584
0.7244	1.2528	550	0.8107	0.7600
0.7123	1.3667	600	0.7905	0.7646
0.6997	1.4806	650	0.7748	0.7678
0.6933	1.5945	700	0.7574	0.7732
0.6822	1.7084	750	0.7393	0.7759
0.6369	1.8223	800	0.7234	0.7806
0.6457	1.9362	850	0.7076	0.7830
0.5368	2.0501	900	0.6982	0.7872
0.5297	2.1640	950	0.6866	0.7885
0.5262	2.2779	1000	0.6760	0.7925
0.5198	2.3918	1050	0.6645	0.7946
0.5167	2.5057	1100	0.6569	0.7968
0.4903	2.6196	1150	0.6523	0.7983
0.4961	2.7335	1200	0.6478	0.7988
0.5140	2.8474	1250	0.6459	0.7996
0.5092	2.9613	1300	0.6452	0.7998
0.5048	3.0	1317	0.6452	0.7996

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 3.6.0
Tokenizers 0.22.1

Task: Text-to-Audio
Library: transformers
Type: qwen2_5_omni
Created: Oct 15, 2025
Updated: Oct 15, 2025