Update README.md

README.md

@@ -127,9 +127,11 @@ tokenizer.push_to_hub(save_to)
 # Model Quality
 TODO
 
-# Peak Memory Usage
+# Memory Usage
 
-|                                  |                |                               |
+Tested on H100
+
+| Memory                           |                |                               |
 |----------------------------------|----------------|-------------------------------|
 |                                  | Qwen3-32B      | Qwen3-32B-float8dq            | 
 | Peak Memory                      | 65.72 GB       | 34.54 GB (-47.44%)            |
@@ -196,7 +198,7 @@ print(f"Peak Memory Usage: {mem:.02f} GB")
 
 # Model Performance
 
-
+Tested on H100
 | Benchmark                        |                |                               |
 |----------------------------------|----------------|-------------------------------|
 |                                  | Qwen3-32B      | Qwen3-32B-float8dq            |