auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -32,10 +32,20 @@ more details, including on how to concatenate multi-part files.
|
|
| 32 |
|:-----|:-----|--------:|:------|
|
| 33 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q2_K.gguf) | i1-Q2_K | 11.4 | IQ3_XXS probably better |
|
| 34 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 11.9 | lower quality |
|
|
|
|
|
|
|
|
|
|
| 35 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_M.gguf) | i1-IQ3_M | 13.6 | |
|
| 36 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 14.8 | IQ3_S probably better |
|
|
|
|
|
|
|
|
|
|
| 37 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 17.6 | optimal size/speed/quality |
|
| 38 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 18.7 | fast, recommended |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 41 |
types (lower is better):
|
|
|
|
| 32 |
|:-----|:-----|--------:|:------|
|
| 33 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q2_K.gguf) | i1-Q2_K | 11.4 | IQ3_XXS probably better |
|
| 34 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 11.9 | lower quality |
|
| 35 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 12.7 | |
|
| 36 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 13.4 | IQ3_XS probably better |
|
| 37 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_S.gguf) | i1-IQ3_S | 13.4 | beats Q3_K* |
|
| 38 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ3_M.gguf) | i1-IQ3_M | 13.6 | |
|
| 39 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 14.8 | IQ3_S probably better |
|
| 40 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 16.0 | IQ3_M probably better |
|
| 41 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 16.5 | |
|
| 42 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_0.gguf) | i1-Q4_0 | 17.5 | fast, low quality |
|
| 43 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 17.6 | optimal size/speed/quality |
|
| 44 |
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 18.7 | fast, recommended |
|
| 45 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q4_1.gguf) | i1-Q4_1 | 19.3 | |
|
| 46 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 21.2 | |
|
| 47 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 21.8 | |
|
| 48 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen3-30B-A3B-i1-GGUF/resolve/main/Qwen3-30B-A3B.i1-Q6_K.gguf) | i1-Q6_K | 25.2 | practically like static Q6_K |
|
| 49 |
|
| 50 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 51 |
types (lower is better):
|