auto-patch README.md
Browse files
README.md
CHANGED
@@ -21,7 +21,7 @@ quantized_by: mradermacher
|
|
21 |
static quants of https://huggingface.co/TencentARC/Mistral_Pro_8B_v0.1
|
22 |
|
23 |
<!-- provided-files -->
|
24 |
-
weighted/imatrix quants
|
25 |
## Usage
|
26 |
|
27 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
@@ -37,9 +37,15 @@ more details, including on how to concatenate multi-part files.
|
|
37 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q2_K.gguf) | Q2_K | 3.5 | |
|
38 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q3_K_S.gguf) | Q3_K_S | 4.0 | |
|
39 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q3_K_M.gguf) | Q3_K_M | 4.5 | lower quality |
|
|
|
|
|
40 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q4_K_S.gguf) | Q4_K_S | 5.2 | fast, recommended |
|
|
|
|
|
|
|
41 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q6_K.gguf) | Q6_K | 7.5 | very good quality |
|
42 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q8_0.gguf) | Q8_0 | 9.7 | fast, best quality |
|
|
|
43 |
|
44 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
45 |
types (lower is better):
|
|
|
21 |
static quants of https://huggingface.co/TencentARC/Mistral_Pro_8B_v0.1
|
22 |
|
23 |
<!-- provided-files -->
|
24 |
+
weighted/imatrix quants are available at https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-i1-GGUF
|
25 |
## Usage
|
26 |
|
27 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
|
37 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q2_K.gguf) | Q2_K | 3.5 | |
|
38 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q3_K_S.gguf) | Q3_K_S | 4.0 | |
|
39 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q3_K_M.gguf) | Q3_K_M | 4.5 | lower quality |
|
40 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q3_K_L.gguf) | Q3_K_L | 4.8 | |
|
41 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.IQ4_XS.gguf) | IQ4_XS | 5.0 | |
|
42 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q4_K_S.gguf) | Q4_K_S | 5.2 | fast, recommended |
|
43 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q4_K_M.gguf) | Q4_K_M | 5.5 | fast, recommended |
|
44 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q5_K_S.gguf) | Q5_K_S | 6.3 | |
|
45 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q5_K_M.gguf) | Q5_K_M | 6.5 | |
|
46 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q6_K.gguf) | Q6_K | 7.5 | very good quality |
|
47 |
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.Q8_0.gguf) | Q8_0 | 9.7 | fast, best quality |
|
48 |
+
| [GGUF](https://huggingface.co/mradermacher/Mistral_Pro_8B_v0.1-GGUF/resolve/main/Mistral_Pro_8B_v0.1.f16.gguf) | f16 | 18.1 | 16 bpw, overkill |
|
49 |
|
50 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
51 |
types (lower is better):
|