Update README.md

README.md

@@ -2,6 +2,13 @@
 license: mit
 ---
 Instruction-tuned model finetuned on WizardLMTeam/WizardLM_evol_instruct_V2_196k
+- Small enough to be run on a phone
+- 124 million parameters
+- Comparable performance to TinyLlama-Chat
+
+We ran some zero-shot tests to compare Lazarus Instruct with the much larger TinyLlama-Chat
+![Alt text](https://huggingface.co/Aclevo/Lazarus-Instruct/blob/main/benchmark.png)
+
 
 ## 🚀 Usage