Update README.md

README.md

@@ -7,10 +7,14 @@ sdk: static
 pinned: false
 ---
 
-# Software-delivered AI Inference
+# Software-Delivered AI Inference
 
 Neural Magic helps developers in accelerating deep learning performance using automated model sparsification technologies and a CPU inference engine.
-Download and run our sparsity-aware inference engine and open source tools for GPU-class performance on CPUs.
+Download our sparsity-aware inference engine and open source tools for GPU-class performance on CPUs. 
 * [DeepSparse](https://github.com/neuralmagic/deepsparse): Inference runtime offering GPU-class performance on CPUs and APIs to integrate ML into your application
 * [SparseML](https://github.com/neuralmagic/sparseml): Libraries for applying sparsification recipes to neural networks with a few lines of code, enabling faster and smaller models
 * [SparseZoo](https://sparsezoo.neuralmagic.com/): Neural network model repository for highly sparse and sparse-quantized models with matching sparsification recipes
+
+
+#### ✨NEW✨ DeepSparse LLMs
+We are pleased to announce our paper on Sparse Finetuning of LLMs, starting with MosaicML's MPT-7b. Check out the [paper], [models](https://sparsezoo.neuralmagic.com/?datasets=gsm8k&ungrouped=true), and [usage](https://research.neuralmagic.com/mpt-sparse-finetuning).