Update README.md
Browse files
README.md
CHANGED
|
@@ -18,6 +18,8 @@ pipeline_tag: image-text-to-text
|
|
| 18 |
library_name: transformers
|
| 19 |
---
|
| 20 |
|
|
|
|
|
|
|
| 21 |
# **Qwen3-VisionCaption-2B-Thinking**
|
| 22 |
|
| 23 |
> **Qwen3-VisionCaption-2B-Thinking** is an abliterated v1.0 variant built upon **Qwen3-VL-2B-Instruct-abliterated-v1**, which originates from the **Qwen3-VL-2B-Instruct** architecture. It is specifically optimized for seamless, high precision image captioning and uncensored visual analysis. The model is engineered for robust caption generation, deep reasoning, and unrestricted descriptive understanding across diverse visual and multimodal contexts.
|
|
@@ -107,4 +109,4 @@ print(output_text)
|
|
| 107 |
* May produce explicit, sensitive, or offensive descriptions depending on visual content.
|
| 108 |
* Not recommended for production environments requiring strict safety controls.
|
| 109 |
* Performance may vary for heavily abstract or synthetic content.
|
| 110 |
-
* Output tone depends on prompt phrasing and detail level settings.
|
|
|
|
| 18 |
library_name: transformers
|
| 19 |
---
|
| 20 |
|
| 21 |
+
|
| 22 |
+
|
| 23 |
# **Qwen3-VisionCaption-2B-Thinking**
|
| 24 |
|
| 25 |
> **Qwen3-VisionCaption-2B-Thinking** is an abliterated v1.0 variant built upon **Qwen3-VL-2B-Instruct-abliterated-v1**, which originates from the **Qwen3-VL-2B-Instruct** architecture. It is specifically optimized for seamless, high precision image captioning and uncensored visual analysis. The model is engineered for robust caption generation, deep reasoning, and unrestricted descriptive understanding across diverse visual and multimodal contexts.
|
|
|
|
| 109 |
* May produce explicit, sensitive, or offensive descriptions depending on visual content.
|
| 110 |
* Not recommended for production environments requiring strict safety controls.
|
| 111 |
* Performance may vary for heavily abstract or synthetic content.
|
| 112 |
+
* Output tone depends on prompt phrasing and detail level settings.
|