mradermacher
commited on
Commit
•
1f4bf4a
1
Parent(s):
d6c9748
auto-patch README.md
Browse files
README.md
CHANGED
@@ -19,7 +19,7 @@ tags:
|
|
19 |
static quants of https://huggingface.co/lodrick-the-lafted/Grafted-Llama2-2x70B
|
20 |
|
21 |
<!-- provided-files -->
|
22 |
-
weighted/imatrix quants
|
23 |
## Usage
|
24 |
|
25 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
@@ -32,7 +32,17 @@ more details, including on how to concatenate multi-part files.
|
|
32 |
|
33 |
| Link | Type | Size/GB | Notes |
|
34 |
|:-----|:-----|--------:|:------|
|
|
|
|
|
|
|
|
|
|
|
35 |
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_S.gguf.part2of2) | Q4_K_S | 71.4 | fast, recommended |
|
|
|
|
|
|
|
|
|
|
|
36 |
|
37 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
38 |
types (lower is better):
|
|
|
19 |
static quants of https://huggingface.co/lodrick-the-lafted/Grafted-Llama2-2x70B
|
20 |
|
21 |
<!-- provided-files -->
|
22 |
+
weighted/imatrix quants are available at https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF
|
23 |
## Usage
|
24 |
|
25 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
|
32 |
|
33 |
| Link | Type | Size/GB | Notes |
|
34 |
|:-----|:-----|--------:|:------|
|
35 |
+
| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q2_K.gguf) | Q2_K | 46.0 | |
|
36 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_S.gguf.part2of2) | Q3_K_S | 54.2 | |
|
37 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_M.gguf.part2of2) | Q3_K_M | 60.2 | lower quality |
|
38 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q3_K_L.gguf.part2of2) | Q3_K_L | 65.3 | |
|
39 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.IQ4_XS.gguf.part2of2) | IQ4_XS | 67.6 | |
|
40 |
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_S.gguf.part2of2) | Q4_K_S | 71.4 | fast, recommended |
|
41 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q4_K_M.gguf.part2of2) | Q4_K_M | 75.7 | fast, recommended |
|
42 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q5_K_S.gguf.part2of2) | Q5_K_S | 86.3 | |
|
43 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q5_K_M.gguf.part2of2) | Q5_K_M | 88.9 | |
|
44 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q6_K.gguf.part3of3) | Q6_K | 102.9 | very good quality |
|
45 |
+
| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q8_0.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q8_0.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF/resolve/main/Grafted-Llama2-2x70B.Q8_0.gguf.part3of3) | Q8_0 | 133.3 | fast, best quality |
|
46 |
|
47 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
48 |
types (lower is better):
|