Update README.md

README.md

@@ -12,7 +12,7 @@ language:
 
 # Meet7 0.6B — Experimental Thinking
 
-A thinking trained variant of [Meet7 0.6B Experimental](https://huggingface.co/Ma7ee7/Meet7_0.6b-experimental), re-enabling Qwen3's built-in chain-of-thought reasoning at inference time.
+A thinking trained variant of [Meet7 0.6B Experimental](https://huggingface.co/Ma7ee7/Meet7_0.6b_Exp), re-enabling Qwen3's built-in chain-of-thought reasoning at inference time.
 
 > **Note:** At 0.6B scale, thinking mode does not improve benchmark performance. (Except over base model which was tested with reasoning.) The model still lacks sufficient capacity to reason coherently across extended thought chains. For best results, use [Meet7 Experimental](https://huggingface.co/Ma7ee7/Meet7_0.6b-experimental) without thinking mode, or [Meet7 0.6B](https://huggingface.co/Ma7ee7/Meet7_0.6b) if BoolQ-style QA is your primary use case.
 
@@ -47,7 +47,7 @@ A thinking trained variant of [Meet7 0.6B Experimental](https://huggingface.co/M
 | Model | Strengths | Weaknesses |
 |-------|-----------|------------|
 | [Meet7 0.6B](https://huggingface.co/Ma7ee7/Meet7_0.6b) | BoolQ (+17.56% vs base) | Less balanced overall |
-| [Meet7 Experimental](https://huggingface.co/Ma7ee7/Meet7_0.6b-experimental) | Best overall balance, wins 4/6 tasks | BoolQ regresses vs Meet7 |
+| [Meet7 Experimental](https://huggingface.co/Ma7ee7/Meet7_0.6b_Exp) | Best overall balance, wins 4/6 tasks | BoolQ regresses vs Meet7 |
 | Meet7 Exp_Thinking *(this model)* | Thinking mode enabled | **(OUR)** Weakest benchmark scores at 0.6B scale |
 
 ## Model Details