Update README.md

README.md

@@ -36,6 +36,8 @@ license: apache-2.0
 
 ## Llamacpp imatrix Quantizations of Devstral-Small-2505 by mistralai
 
+### Created using Unsloth's conversion (huge thanks for uploading it) from here: https://huggingface.co/unsloth/Devstral-Small-2505
+
 Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b5432">b5432</a> for quantization.
 
 Original model: https://huggingface.co/mistralai/Devstral-Small-2505
@@ -46,6 +48,12 @@ Run them in [LM Studio](https://lmstudio.ai/)
 
 Run them directly with [llama.cpp](https://github.com/ggerganov/llama.cpp), or any other llama.cpp based project
 
+## Vision support thanks to ngxson
+
+ngxson has shown that the original mmproj from Mistral Small 3.1 works seamlessly with Devstral Small, so the mmproj file is here for vision use!
+
+Note this is not official, and may not work perfectly, but in practice seems functional!
+
 ## Prompt format
 
 ```