Update README.md

README.md

@@ -14,4 +14,32 @@ tags:
 
 # **DeepCaption-VLA-7B-AIO-GGUF**
 
-> DeepCaption-VLA-7B from prithivMLmods is a 7B-parameter vision-language model fine-tuned from Qwen2.5-VL-7B-Instruct, specialized for image captioning and Vision Language Attribution (VLA) that generates precise, attribute-rich descriptions of visual properties, object attributes, scene details, colors, environments, moods, and actions across general, artistic, technical, abstract, or low-context images in diverse aspect ratios (wide, tall, square, irregular). Trained on curated datasets like blip3o-caption-mini-arrow, Caption3o-Opt-v3/v2, Caltech101 attributes, and private domain-specific sources to emphasize object-attribute alignment and descriptive fluency, it supports variational detail control—from concise summaries to fine-grained attributions—via structured outputs including captions, comma-separated attributes, and {class_name==core_theme} syntax, primarily in English with multilingual prompt adaptability. Ideal for research/dataset creation, object detection annotation, scene understanding, and creative applications using Transformers/Qwen2VLForConditionalGeneration inference, it handles non-standard visuals robustly but may over-attribute ambiguous content or vary by prompt phrasing.
+> DeepCaption-VLA-7B from prithivMLmods is a 7B-parameter vision-language model fine-tuned from Qwen2.5-VL-7B-Instruct, specialized for image captioning and Vision Language Attribution (VLA) that generates precise, attribute-rich descriptions of visual properties, object attributes, scene details, colors, environments, moods, and actions across general, artistic, technical, abstract, or low-context images in diverse aspect ratios (wide, tall, square, irregular). Trained on curated datasets like blip3o-caption-mini-arrow, Caption3o-Opt-v3/v2, Caltech101 attributes, and private domain-specific sources to emphasize object-attribute alignment and descriptive fluency, it supports variational detail control—from concise summaries to fine-grained attributions—via structured outputs including captions, comma-separated attributes, and {class_name==core_theme} syntax, primarily in English with multilingual prompt adaptability. Ideal for research/dataset creation, object detection annotation, scene understanding, and creative applications using Transformers/Qwen2VLForConditionalGeneration inference, it handles non-standard visuals robustly but may over-attribute ambiguous content or vary by prompt phrasing.
+
+## DeepCaption-VLA-7B [GGUF]
+
+| File Name | Quant Type | File Size | File Link |
+| - | - | - | - |
+| DeepCaption-VLA-7B.IQ4_XS.gguf | IQ4_XS | 4.25 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.IQ4_XS.gguf) |
+| DeepCaption-VLA-7B.Q2_K.gguf | Q2_K | 3.02 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q2_K.gguf) |
+| DeepCaption-VLA-7B.Q3_K_L.gguf | Q3_K_L | 4.09 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q3_K_L.gguf) |
+| DeepCaption-VLA-7B.Q3_K_M.gguf | Q3_K_M | 3.81 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q3_K_M.gguf) |
+| DeepCaption-VLA-7B.Q3_K_S.gguf | Q3_K_S | 3.49 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q3_K_S.gguf) |
+| DeepCaption-VLA-7B.Q4_K_M.gguf | Q4_K_M | 4.68 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q4_K_M.gguf) |
+| DeepCaption-VLA-7B.Q4_K_S.gguf | Q4_K_S | 4.46 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q4_K_S.gguf) |
+| DeepCaption-VLA-7B.Q5_K_M.gguf | Q5_K_M | 5.44 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q5_K_M.gguf) |
+| DeepCaption-VLA-7B.Q5_K_S.gguf | Q5_K_S | 5.32 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q5_K_S.gguf) |
+| DeepCaption-VLA-7B.Q6_K.gguf | Q6_K | 6.25 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q6_K.gguf) |
+| DeepCaption-VLA-7B.Q8_0.gguf | Q8_0 | 8.1 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.Q8_0.gguf) |
+| DeepCaption-VLA-7B.f16.gguf | F16 | 15.2 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.f16.gguf) |
+| DeepCaption-VLA-7B.mmproj-Q8_0.gguf | mmproj-Q8_0 | 856 MB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.mmproj-Q8_0.gguf) |
+| DeepCaption-VLA-7B.mmproj-f16.gguf | mmproj-f16 | 1.35 GB | [Download](https://huggingface.co/prithivMLmods/DeepCaption-VLA-7B-AIO-GGUF/blob/main/DeepCaption-VLA-7B.mmproj-f16.gguf) |
+
+## Quants Usage 
+
+(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
+
+Here is a handy graph by ikawrakow comparing some lower-quality quant
+types (lower is better):
+
+![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)