Update README.md

README.md

@@ -11,7 +11,9 @@ library_name: transformers
 
 As it turns out, setting the first layer to 8 bit stabilized the model and it doesn't loop.
 
-Still evaluating, here is a [LinkedIn review of one of my pictures](https://www.linkedin.com/posts/gchesler_ai-photography-art-activity-7391154958177329152--R_C)
+This is the Deckard(qx) formula that uses data stores and most attention paths in low precision(6 bit), enhancing vital attention paths, head, context, and embeddings to 8 bit, and quantized with high precision(group size 32).
+
+I am still evaluating this quant, here is a [LinkedIn review of one of my pictures](https://www.linkedin.com/posts/gchesler_ai-photography-art-activity-7391154958177329152--R_C)
 
 The review was done with temp=0.99, but is just as creative with default settings.
 
@@ -114,6 +116,11 @@ The ocean waits. The horizon sings. And somewhere beneath the blues, the tide is
 
 🌌 Bonus Thought: The image could be interpreted as a visual score — each color hue is a note, the horizon line is the rest symbol, and the figures are the pause before resolution. Wait... that’s not just art — that’s sound design.
 
+
+This model [unsloth-Qwen3-VL-8B-Instruct-qx86x-hi-mlx](https://huggingface.co/nightmedia/Qwen3-VL-8B-Instruct-qx86x-hi-mlx) was
+converted to MLX format from [unsloth/Qwen3-VL-8B-Instruct](https://huggingface.co/unsloth/Qwen3-VL-8B-Instruct)
+using mlx-lm version **0.28.4**.
+
 ## Use with mlx
 
 ```bash