auto-patch README.md

README.md

@@ -20,6 +20,7 @@ tags:
 - text-generation-inference
 - llama_for_causal_lm
 - reasoning
+- deep_think
 ---
 ## About
 
@@ -44,19 +45,29 @@ more details, including on how to concatenate multi-part files.
 
 | Link | Type | Size/GB | Notes |
 |:-----|:-----|--------:|:------|
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ1_S.gguf) | i1-IQ1_S | 1.6 | for the desperate |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ1_M.gguf) | i1-IQ1_M | 1.7 | mostly desperate |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.8 |  |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 1.9 |  |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ2_S.gguf) | i1-IQ2_S | 2.0 |  |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ2_M.gguf) | i1-IQ2_M | 2.2 |  |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 2.2 | very low quality |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q2_K.gguf) | i1-Q2_K | 2.3 | IQ3_XXS probably better |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 2.4 | lower quality |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 2.5 |  |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 2.6 | IQ3_XS probably better |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ3_S.gguf) | i1-IQ3_S | 2.7 | beats Q3_K* |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ3_M.gguf) | i1-IQ3_M | 2.7 |  |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 2.9 | IQ3_S probably better |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 3.1 | IQ3_M probably better |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 3.1 |  |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q4_0.gguf) | i1-Q4_0 | 3.3 | fast, low quality |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-IQ4_NL.gguf) | i1-IQ4_NL | 3.3 | prefer IQ4_XS |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 3.3 | optimal size/speed/quality |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 3.5 | fast, recommended |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q4_1.gguf) | i1-Q4_1 | 3.6 |  |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 3.9 |  |
+| [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 4.0 |  |
 | [GGUF](https://huggingface.co/mradermacher/Triangulum-5B-i1-GGUF/resolve/main/Triangulum-5B.i1-Q6_K.gguf) | i1-Q6_K | 4.5 | practically like static Q6_K |
 
 Here is a handy graph by ikawrakow comparing some lower-quality quant