|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- zh |
|
|
- en |
|
|
base_model: |
|
|
- huihui-ai/Qwen2.5-VL-7B-Instruct-abliterated |
|
|
pipeline_tag: image-text-to-text |
|
|
tags: |
|
|
- qwen2_5_vl |
|
|
- multimodal |
|
|
- abliterated |
|
|
- uncensored |
|
|
- text-generation-inference |
|
|
--- |
|
|
|
|
|
Quantized gguf file from https://huggingface.co/huihui-ai/Qwen2.5-VL-7B-Instruct-abliterated |
|
|
|
|
|
Using "--leave-output-tensor" in quantizing to keep output layer precision at FP16. |
|
|
|
|
|
LM Studio is recommended to deploy it. |
|
|
|
|
|
**Runtime environment should be upgraded to >= 1.29.0(beta).** |
|
|
|
|
|
imatrix.dat is from mradermacher/Qwen2.5-VL-7B-Instruct-abliterated-i1-GGUF |
