Commit
•
6404b70
1
Parent(s):
23c0b43
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -18,7 +18,6 @@ quantized_by: mradermacher
|
|
| 18 |
weighted/imatrix quants of https://huggingface.co/Qwen/Qwen2.5-72B
|
| 19 |
|
| 20 |
<!-- provided-files -->
|
| 21 |
-
static quants are available at https://huggingface.co/mradermacher/Qwen2.5-72B-GGUF
|
| 22 |
## Usage
|
| 23 |
|
| 24 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
@@ -37,6 +36,7 @@ more details, including on how to concatenate multi-part files.
|
|
| 37 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 27.2 | |
|
| 38 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_S.gguf) | i1-IQ2_S | 28.0 | |
|
| 39 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_M.gguf) | i1-IQ2_M | 29.4 | |
|
|
|
|
| 40 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q2_K.gguf) | i1-Q2_K | 29.9 | IQ3_XXS probably better |
|
| 41 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 31.9 | lower quality |
|
| 42 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 32.9 | |
|
|
@@ -46,11 +46,18 @@ more details, including on how to concatenate multi-part files.
|
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 37.8 | IQ3_S probably better |
|
| 47 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 39.6 | IQ3_M probably better |
|
| 48 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 39.8 | |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 49 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_0.gguf) | i1-Q4_0 | 41.5 | fast, low quality |
|
| 50 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 44.0 | optimal size/speed/quality |
|
|
|
|
| 51 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 47.5 | fast, recommended |
|
|
|
|
| 52 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 51.5 | |
|
| 53 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 54.5 | |
|
|
|
|
| 54 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 64.4 | practically like static Q6_K |
|
| 55 |
|
| 56 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
|
|
| 18 |
weighted/imatrix quants of https://huggingface.co/Qwen/Qwen2.5-72B
|
| 19 |
|
| 20 |
<!-- provided-files -->
|
|
|
|
| 21 |
## Usage
|
| 22 |
|
| 23 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
|
|
| 36 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 27.2 | |
|
| 37 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_S.gguf) | i1-IQ2_S | 28.0 | |
|
| 38 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ2_M.gguf) | i1-IQ2_M | 29.4 | |
|
| 39 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 29.7 | very low quality |
|
| 40 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q2_K.gguf) | i1-Q2_K | 29.9 | IQ3_XXS probably better |
|
| 41 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 31.9 | lower quality |
|
| 42 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 32.9 | |
|
|
|
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 37.8 | IQ3_S probably better |
|
| 47 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 39.6 | IQ3_M probably better |
|
| 48 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 39.8 | |
|
| 49 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 41.3 | fast on arm, low quality |
|
| 50 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 41.3 | fast on arm+i8mm, low quality |
|
| 51 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 41.3 | fast on arm+sve, low quality |
|
| 52 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-IQ4_NL.gguf) | i1-IQ4_NL | 41.4 | prefer IQ4_XS |
|
| 53 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_0.gguf) | i1-Q4_0 | 41.5 | fast, low quality |
|
| 54 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 44.0 | optimal size/speed/quality |
|
| 55 |
+
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_1.gguf) | i1-Q4_1 | 45.8 | |
|
| 56 |
| [GGUF](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 47.5 | fast, recommended |
|
| 57 |
+
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_0.gguf.part2of2) | i1-Q5_0 | 50.4 | |
|
| 58 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 51.5 | |
|
| 59 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 54.5 | |
|
| 60 |
+
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_1.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q5_1.gguf.part2of2) | i1-Q5_1 | 54.7 | |
|
| 61 |
| [PART 1](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-72B-i1-GGUF/resolve/main/Qwen2.5-72B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 64.4 | practically like static Q6_K |
|
| 62 |
|
| 63 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|