auto-patch README.md

README.md

@@ -73,10 +73,13 @@ more details, including on how to concatenate multi-part files.
 | Link | Type | Size/GB | Notes |
 |:-----|:-----|--------:|:------|
 | [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.imatrix.gguf) | imatrix | 0.3 | imatrix file (for creating your own qwuants) |
+| [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-IQ2_M.gguf) | i1-IQ2_M | 17.6 |  |
 | [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-Q2_K.gguf) | i1-Q2_K | 19.5 | IQ3_XXS probably better |
 | [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 20.6 | lower quality |
 | [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-IQ3_M.gguf) | i1-IQ3_M | 23.4 |  |
+| [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-Q3_K_M.gguf) | i1-Q3_K_M | 25.5 | IQ3_S probably better |
 | [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-Q4_K_S.gguf) | i1-Q4_K_S | 30.3 | optimal size/speed/quality |
+| [GGUF](https://huggingface.co/mradermacher/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL-i1-GGUF/resolve/main/Qwen3-Yoyo-V3-54B-A3B-Thinking-TOTAL-RECALL.i1-Q4_K_M.gguf) | i1-Q4_K_M | 32.2 | fast, recommended |
 
 Here is a handy graph by ikawrakow comparing some lower-quality quant
 types (lower is better):