Add Key Results summary block

README.md

@@ -40,7 +40,16 @@ Dystrio Sculpt produces dense compiled variants of existing models that:
 - require no custom kernels
 - load with standard HuggingFace Transformers
 
+## Key Results
 
+Compared to **mistralai/Mistral-7B-v0.1** baseline on an **A100 80GB**:
+
+- **Weights memory:** **-11% (Conservative)** / **-23% (Balanced)**
+- **RAG latency (TTFT p95):** **-7% / -14%**
+- **Decode throughput:** ~flat
+- **No runtime changes:** no custom kernels, no new ops, standard `transformers` loading
+
+> Notes: TTFT includes prefill + first decode step. “Weights memory” is computed from parameter sizes (GiB) and is workload-independent.
 
 
 ## Benchmark Results