| ### Note: DO NOT use quantized model or quantization_bit when merging lora adapters | |
| ### model | |
| # model_name_or_path: Qwen/Qwen2.5-VL-7B-Instruct | |
| model_name_or_path: llava-hf/llama3-llava-next-8b-hf | |
| adapter_name_or_path: saves/mol-instruct-llava3-next/checkpoint-1000 | |
| template: llava_next | |
| trust_remote_code: true | |
| ### export | |
| export_dir: saves/merged_models/mol-instruct-llava3-next-checkpoint-1000 | |
| export_size: 5 | |
| export_device: cpu # choices: [cpu, auto] | |
| export_legacy_format: false |