Commit
•
2e4a77d
1
Parent(s):
26b376c
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -15,7 +15,6 @@ tags:
|
|
| 15 |
|
| 16 |
weighted/imatrix quants of https://huggingface.co/MarsupialAI/KitchenSink_103b
|
| 17 |
|
| 18 |
-
|
| 19 |
<!-- provided-files -->
|
| 20 |
static quants are available at https://huggingface.co/mradermacher/KitchenSink_103b-GGUF
|
| 21 |
## Usage
|
|
@@ -31,8 +30,12 @@ more details, including on how to concatenate multi-part files.
|
|
| 31 |
| Link | Type | Size/GB | Notes |
|
| 32 |
|:-----|:-----|--------:|:------|
|
| 33 |
| [GGUF](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q2_K.gguf) | i1-Q2_K | 38.3 | IQ3_XXS probably better |
|
|
|
|
| 34 |
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 50.0 | IQ3_S probably better |
|
|
|
|
| 35 |
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 59.0 | optimal size/speed/quality |
|
|
|
|
|
|
|
| 36 |
|
| 37 |
|
| 38 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
|
|
| 15 |
|
| 16 |
weighted/imatrix quants of https://huggingface.co/MarsupialAI/KitchenSink_103b
|
| 17 |
|
|
|
|
| 18 |
<!-- provided-files -->
|
| 19 |
static quants are available at https://huggingface.co/mradermacher/KitchenSink_103b-GGUF
|
| 20 |
## Usage
|
|
|
|
| 30 |
| Link | Type | Size/GB | Notes |
|
| 31 |
|:-----|:-----|--------:|:------|
|
| 32 |
| [GGUF](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q2_K.gguf) | i1-Q2_K | 38.3 | IQ3_XXS probably better |
|
| 33 |
+
| [GGUF](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_S.gguf) | i1-Q3_K_S | 44.9 | IQ3_XS probably better |
|
| 34 |
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 50.0 | IQ3_S probably better |
|
| 35 |
+
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 54.5 | IQ3_M probably better |
|
| 36 |
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 59.0 | optimal size/speed/quality |
|
| 37 |
+
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 62.3 | fast, recommended |
|
| 38 |
+
| [PART 1](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/KitchenSink_103b-i1-GGUF/resolve/main/KitchenSink_103b.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 85.1 | practically like static Q6_K |
|
| 39 |
|
| 40 |
|
| 41 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|