mradermacher
commited on
Commit
•
c1a8eb6
1
Parent(s):
7d166b5
auto-patch README.md
Browse files
README.md
CHANGED
@@ -29,16 +29,27 @@ more details, including on how to concatenate multi-part files.
|
|
29 |
| Link | Type | Size/GB | Notes |
|
30 |
|:-----|:-----|--------:|:------|
|
31 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ1_S.gguf) | i1-IQ1_S | 25.7 | for the desperate |
|
|
|
32 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 32.2 | |
|
33 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 35.8 | |
|
|
|
|
|
34 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q2_K.gguf) | i1-Q2_K | 44.6 | IQ3_XXS probably better |
|
35 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 47.3 | lower quality |
|
|
|
36 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_XS.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_XS.gguf.split-ab) | i1-Q3_K_XS | 49.3 | |
|
|
|
37 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_S.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_S.gguf.split-ab) | i1-Q3_K_S | 52.2 | IQ3_XS probably better |
|
|
|
38 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_M.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_M.gguf.split-ab) | i1-Q3_K_M | 58.2 | IQ3_S probably better |
|
39 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_L.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_L.gguf.split-ab) | i1-Q3_K_L | 63.4 | IQ3_M probably better |
|
|
|
|
|
40 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_S.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_S.gguf.split-ab) | i1-Q4_K_S | 68.7 | optimal size/speed/quality |
|
41 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_M.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_M.gguf.split-ab) | i1-Q4_K_M | 72.6 | fast, recommended |
|
|
|
|
|
|
|
42 |
|
43 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
44 |
types (lower is better):
|
|
|
29 |
| Link | Type | Size/GB | Notes |
|
30 |
|:-----|:-----|--------:|:------|
|
31 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ1_S.gguf) | i1-IQ1_S | 25.7 | for the desperate |
|
32 |
+
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ1_M.gguf) | i1-IQ1_M | 27.8 | mostly desperate |
|
33 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 32.2 | |
|
34 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 35.8 | |
|
35 |
+
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_S.gguf) | i1-IQ2_S | 37.2 | |
|
36 |
+
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ2_M.gguf) | i1-IQ2_M | 40.5 | |
|
37 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q2_K.gguf) | i1-Q2_K | 44.6 | IQ3_XXS probably better |
|
38 |
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 47.3 | lower quality |
|
39 |
+
| [GGUF](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_XS.gguf) | i1-IQ3_XS | 49.3 | |
|
40 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_XS.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_XS.gguf.split-ab) | i1-Q3_K_XS | 49.3 | |
|
41 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 52.1 | beats Q3_K* |
|
42 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_S.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_S.gguf.split-ab) | i1-Q3_K_S | 52.2 | IQ3_XS probably better |
|
43 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 53.8 | |
|
44 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_M.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_M.gguf.split-ab) | i1-Q3_K_M | 58.2 | IQ3_S probably better |
|
45 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_L.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q3_K_L.gguf.split-ab) | i1-Q3_K_L | 63.4 | IQ3_M probably better |
|
46 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 64.3 | |
|
47 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_0.gguf.part2of2) | i1-Q4_0 | 68.1 | fast, low quality |
|
48 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_S.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_S.gguf.split-ab) | i1-Q4_K_S | 68.7 | optimal size/speed/quality |
|
49 |
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_M.gguf.split-aa) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q4_K_M.gguf.split-ab) | i1-Q4_K_M | 72.6 | fast, recommended |
|
50 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 82.9 | |
|
51 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 85.1 | |
|
52 |
+
| [PART 1](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/MegaDolphin-120b-i1-GGUF/resolve/main/MegaDolphin-120b.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 98.8 | practically like static Q6_K |
|
53 |
|
54 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
55 |
types (lower is better):
|