Update README.md

README.md

@@ -50,11 +50,21 @@ Mixtral 8x7B is a generative Sparse Mixture of Experts (SMoE) model designed to
 
 ### Core Library
 
-Mixtral 8x7B Instruct can be deployed using `vLLM` or `transformers`. Current support focuses on Hugging Face `transformers` for initial integrations.
+Mixtral 8x7B Instruct is supported by multiple libraries to ensure flexibility for deployment and development. The primary frameworks include:
 
-**Primary Framework**: `transformers`  
-**Alternate Framework**: `vLLM` (for specialized inference optimizations)  
-**Model Availability**: Source weights and pre-converted formats are available under [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0).
+- **Primary Framework**: `llama.cpp`  
+- **Alternate Frameworks**:
+  - `transformers` for initial integration into Hugging Face's ecosystem.
+  - `vLLM` for highly optimized inference with low-latency serving.
+
+You can access the model components and libraries here:
+
+- **Model Base**: [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1)  
+- **Common Utilities**: [mistralai/mistral-common](https://github.com/mistralai/mistral-common)  
+- **Inference Optimization**: [mistralai/mistral-inference](https://github.com/mistralai/mistral-inference)  
+- **Quantization Support**: [ggerganov/llama.cpp](https://github.com/ggerganov/llama.cpp)  
+
+These resources provide a complete ecosystem for deployment, fine-tuning, and scaling sparse mixture models.
 
 ### Safety and Responsible Use