Update README.md

README.md

@@ -15,9 +15,8 @@ language:
 pipeline_tag: text-generation
 ---
 
-<img src="https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/EgsjPDWd37LjAtamiICxk.png" width="300" height="300" alt="image/png">
+<img src="https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/EgsjPDWd37LjAtamiICxk.png" width="480" height="480" alt="image/png">
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/EgsjPDWd37LjAtamiICxk.png)
 
 ### Disclaimer
 This model is a base model which received aggressive pruning and knowledge distillation. To make it usable for your individual application it must we finetuned.
@@ -66,7 +65,8 @@ Up to 40 % parameter reduction (24 B → 15 B) delivers 2× lower TTFT
 | Tokens / s | 579 | **812** | +40% |
 
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/4rDhaeC-1GMj6KWbB27f9.png)
+
+<img src="https://cdn-uploads.huggingface.co/production/uploads/645ded34a45b4182d7f5c385/4rDhaeC-1GMj6KWbB27f9.png" width="300" height="300" alt="image/png">
 
 ### Training scalability (distillation run, MI300A cluster)