Update README.md

README.md

@@ -13,7 +13,23 @@ tags:
 
 This repository hosts quantized GGUF files for **Loyal-Macaroni-Maid-7B**, a 7B model aimed at having engaging RP with solid character card adherence and being a smart cookie at the same time.
 
-In my limited testing, it's a great RP model suitable for RP/ERP with sharp reasoning skills for a 7B. I expect it to both benchmark well and be a very suitable model for general use.
+In my limited testing, it's a great RP model suitable for RP/ERP with sharp reasoning skills for a 7B.
+
+On benchmarks like MT-Bench, this RP model scores shockingly well.
+
+| Model                  | MT Bench | MMLU |
+|------------------------|----------|------|
+| GPT-4-Turbo            | 9.32     |      |
+| GPT-4                  | 8.99     | 86.4 |
+| Starling-7B            | 8.09     | 63.9 |
+| Claude-2               | 8.06     | 78.5 |
+| Loyal-Macaroni-Maid-7B | 7.95     | ~64.9|
+| GPT-3.5-Turbo          | 7.94     | 70   |
+| Claude-1               | 7.9      | 77   |
+| Openchat-3.5           | 7.81     | 64.3 |
+| Zephyr-7B-beta         | 7.34     | 61.4 |
+| Llama-2-70b-chat-hf    | 6.86     | 63   |
+| Neural-chat-7b-v3-1    | 6.84     | 62.4 | 
 
 <!-- description end -->
 <!-- prompt-template start -->