mradermacher
commited on
Commit
•
8671864
1
Parent(s):
ada9500
auto-patch README.md
Browse files
README.md
CHANGED
@@ -29,6 +29,17 @@ more details, including on how to concatenate multi-part files.
|
|
29 |
|
30 |
| Link | Type | Size/GB | Notes |
|
31 |
|:-----|:-----|--------:|:------|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part4of4) | Q8_0 | 149.5 | fast, best quality |
|
33 |
|
34 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
|
29 |
|
30 |
| Link | Type | Size/GB | Notes |
|
31 |
|:-----|:-----|--------:|:------|
|
32 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q2_K.gguf.part2of2) | Q2_K | 52.2 | |
|
33 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.IQ3_S.gguf.part2of2) | IQ3_S | 61.6 | beats Q3_K* |
|
34 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_S.gguf.part2of2) | Q3_K_S | 61.6 | |
|
35 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.IQ3_M.gguf.part2of2) | IQ3_M | 64.6 | |
|
36 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_M.gguf.part2of2) | Q3_K_M | 67.9 | lower quality |
|
37 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q3_K_L.gguf.part2of2) | Q3_K_L | 72.7 | |
|
38 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q4_K_S.gguf.part2of2) | Q4_K_S | 80.6 | fast, recommended |
|
39 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q4_K_M.gguf.part2of2) | Q4_K_M | 85.7 | fast, recommended |
|
40 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q5_K_S.gguf.part2of2) | Q5_K_S | 97.1 | |
|
41 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q5_K_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q5_K_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q5_K_M.gguf.part3of3) | Q5_K_M | 100.1 | |
|
42 |
+
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q6_K.gguf.part3of3) | Q6_K | 115.6 | very good quality |
|
43 |
| [PART 1](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/mixtral-8x22B-v0.3-GGUF/resolve/main/mixtral-8x22B-v0.3.Q8_0.gguf.part4of4) | Q8_0 | 149.5 | fast, best quality |
|
44 |
|
45 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|