Commit
•
bd588ff
1
Parent(s):
423b6df
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -31,15 +31,22 @@ more details, including on how to concatenate multi-part files.
|
|
| 31 |
|
| 32 |
| Link | Type | Size/GB | Notes |
|
| 33 |
|:-----|:-----|--------:|:------|
|
|
|
|
| 34 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ2_M.gguf) | i1-IQ2_M | 29.4 | |
|
| 35 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q2_K.gguf) | i1-Q2_K | 29.9 | IQ3_XXS probably better |
|
| 36 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 31.9 | lower quality |
|
|
|
|
|
|
|
| 37 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_S.gguf) | i1-Q3_K_S | 34.6 | IQ3_XS probably better |
|
|
|
|
| 38 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_M.gguf) | i1-Q3_K_M | 37.8 | IQ3_S probably better |
|
| 39 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_L.gguf) | i1-Q3_K_L | 39.6 | IQ3_M probably better |
|
| 40 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ4_XS.gguf) | i1-IQ4_XS | 39.8 | |
|
|
|
|
| 41 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q4_K_S.gguf) | i1-Q4_K_S | 44.0 | optimal size/speed/quality |
|
| 42 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q4_K_M.gguf) | i1-Q4_K_M | 47.5 | fast, recommended |
|
|
|
|
|
|
|
| 43 |
| [PART 1](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 64.4 | practically like static Q6_K |
|
| 44 |
|
| 45 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
|
|
| 31 |
|
| 32 |
| Link | Type | Size/GB | Notes |
|
| 33 |
|:-----|:-----|--------:|:------|
|
| 34 |
+
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 25.6 | |
|
| 35 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ2_M.gguf) | i1-IQ2_M | 29.4 | |
|
| 36 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q2_K.gguf) | i1-Q2_K | 29.9 | IQ3_XXS probably better |
|
| 37 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 31.9 | lower quality |
|
| 38 |
+
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ3_XS.gguf) | i1-IQ3_XS | 32.9 | |
|
| 39 |
+
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ3_S.gguf) | i1-IQ3_S | 34.6 | beats Q3_K* |
|
| 40 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_S.gguf) | i1-Q3_K_S | 34.6 | IQ3_XS probably better |
|
| 41 |
+
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ3_M.gguf) | i1-IQ3_M | 35.6 | |
|
| 42 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_M.gguf) | i1-Q3_K_M | 37.8 | IQ3_S probably better |
|
| 43 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q3_K_L.gguf) | i1-Q3_K_L | 39.6 | IQ3_M probably better |
|
| 44 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-IQ4_XS.gguf) | i1-IQ4_XS | 39.8 | |
|
| 45 |
+
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q4_0.gguf) | i1-Q4_0 | 41.5 | fast, low quality |
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q4_K_S.gguf) | i1-Q4_K_S | 44.0 | optimal size/speed/quality |
|
| 47 |
| [GGUF](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q4_K_M.gguf) | i1-Q4_K_M | 47.5 | fast, recommended |
|
| 48 |
+
| [PART 1](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 51.5 | |
|
| 49 |
+
| [PART 1](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 54.5 | |
|
| 50 |
| [PART 1](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/quill-72b-instruct-i1-GGUF/resolve/main/quill-72b-instruct.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 64.4 | practically like static Q6_K |
|
| 51 |
|
| 52 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|