Update README.md

README.md

@@ -20,6 +20,8 @@ base_model:
 
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65b19c1b098c85365af5a83e/sF7RDZA7lFYOmGy4bGy1s.png)
 
+[imat quants](https://huggingface.co/mradermacher/Qwentile2.5-32B-Instruct-i1-GGUF)
+
 # Qwentile 2.5 32B Instruct
 
 Qwentile 2.5 32B Instruct is a *normalized denoised fourier interpolation* of the following models:
@@ -47,7 +49,9 @@ The first model worked pretty well, but I got a sense that the balances could be
 
 ### Initial Results
 
-It's a little crispier than Awqward, but does generate stable output. Since it is based on Qwen2.5 base instead of instruct, maybe it can score over zero on the math leaderboard?
+It's a little crispier than Awqward, but does generate stable output. Since it is based on Qwen2.5 base instead of instruct, and not only did it not fail the math test, it scores with models twice it's size:
+
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/65b19c1b098c85365af5a83e/wsejMg90Md-Y_13E-osGn.png)
 
 ## Citation