Update README.md

README.md

@@ -12,7 +12,7 @@ Developed by Deep Learning Efficiency Research (DLER) team at NVIDIA Research.
 - Fuse attention heads and SSM heads within the same layer, offering parallel and complementary processing of the same inputs
 
 <div align="center">
-<img src="https://huggingface.co/nvidia/Hymba-1.5B/resolve/main/images/module.png" alt="Hymba Module" width="600">
+<img src="https://huggingface.co/nvidia/Hymba-1.5B-Instruct/resolve/main/images/module.png" alt="Hymba Module" width="600">
 </div>
 
 - Introduce meta tokens that are prepended to the input sequences and interact with all subsequent tokens, thus storing important information and alleviating the burden of "forced-to-attend" in attention
@@ -20,7 +20,7 @@ Developed by Deep Learning Efficiency Research (DLER) team at NVIDIA Research.
 - Integrate with cross-layer KV sharing and global-local attention to further boost memory and computation efficiency
 
 <div align="center">
-<img src="https://huggingface.co/nvidia/Hymba-1.5B/resolve/main/images/macro_arch.png" alt="Hymba Model" width="600">
+<img src="https://huggingface.co/nvidia/Hymba-1.5B-Instruct/resolve/main/images/macro_arch.png" alt="Hymba Model" width="600">
 </div>
 
 
@@ -32,7 +32,7 @@ Developed by Deep Learning Efficiency Research (DLER) team at NVIDIA Research.
 </div>
 
 
-- Hymba-1.5B-Instruct: Outperform all sub-2B public models.
+- Hymba-1.5B-Instruct: Outperform SOTA small LMs.
 
 
 <div align="center">