Update README.md

README.md

@@ -30,10 +30,9 @@ quantized_by: nimishchaudhari
 A high-quality quantized version of [mistralai/Magistral-Small-2509](https://huggingface.co/mistralai/Magistral-Small-2509) using [ik_llama.cpp](https://github.com/ikawrakow/ik_llama.cpp) with importance matrix optimization and custom tensor-specific quantization.
 
 ## Model Details
-- **Base Model:** unsloth/Magistral-Small-2509
+- **Base Model:** mistralai/Magistral-Small-2509
 - **Quantization Method:** Custom IQ4_KS recipe with importance matrix
-- **Target VRAM:** ~7-8GB for full offload
-- **Context Length:** Up to 32K tokens
+- **Target VRAM:** 12GB for full offload
 - **Quantization Tool:** ik_llama.cpp with imatrix calibration
 
 ## Benchmark Results