qwen3-dfm-agent-v1

This model is a fine-tuned version of Qwen/Qwen3-VL-4B-Instruct on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3305

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: 0.0002
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 1200
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.9912	0.9926	100	0.9961
0.7309	1.9826	200	0.7562
0.4736	2.9727	300	0.5362
0.3305	3.9628	400	0.4293
0.2611	4.9529	500	0.3548
0.2274	5.9429	600	0.3313
0.2155	6.9330	700	0.3190
0.1956	7.9231	800	0.3200
0.1945	8.9132	900	0.3195
0.1886	9.9032	1000	0.3215
0.1781	10.8933	1100	0.3233
0.1778	11.8834	1200	0.3305

Framework versions

• PEFT 0.18.0
• Transformers 4.57.2
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1