Commit
•
ccda4ec
1
Parent(s):
121a49e
Update README.md
Browse files
README.md
CHANGED
@@ -51,14 +51,14 @@ They are also compatible with many third party UIs and libraries - please see th
|
|
51 |
|
52 |
| Name | Quant method | Bits | Size | Use case |
|
53 |
| ---- | ---- | ---- | ---- | ----- |
|
54 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q2_K.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q2_k.gguf) | Q2_K | 2 |
|
55 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q3_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q3_k_m.gguf) | Q3_K_M | 3 | 3.
|
56 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q4_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q4_0.gguf) | Q4_0 | 4 |
|
57 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q4_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q4_k_m.gguf) | Q4_K_M | 4 | 4.
|
58 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q5_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q5_0.gguf) | Q5_0 | 5 |
|
59 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q5_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q5_k_m.gguf) | Q5_K_M | 5 |
|
60 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q6_K.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q6_k.gguf) | Q6_K | 6 | 5.
|
61 |
-
| [Merak-7B-v4-PROTOTYPE6-model-Q8_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q8_0.gguf) | Q8_0 | 8 | 7.
|
62 |
|
63 |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
|
64 |
|
|
|
51 |
|
52 |
| Name | Quant method | Bits | Size | Use case |
|
53 |
| ---- | ---- | ---- | ---- | ----- |
|
54 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q2_K.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q2_k.gguf) | Q2_K | 2 | 3.08 GB| smallest, significant quality loss - not recommended for most purposes |
|
55 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q3_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q3_k_m.gguf) | Q3_K_M | 3 | 3.52 GB| very small, high quality loss |
|
56 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q4_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q4_0.gguf) | Q4_0 | 4 | 4.11 GB| legacy; small, very high quality loss - prefer using Q3_K_M |
|
57 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q4_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q4_k_m.gguf) | Q4_K_M | 4 | 4.37 GB| medium, balanced quality - recommended |
|
58 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q5_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q5_0.gguf) | Q5_0 | 5 | 5 GB| legacy; medium, balanced quality - prefer using Q4_K_M |
|
59 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q5_K_M.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q5_k_m.gguf) | Q5_K_M | 5 | 5.13 GB| large, very low quality loss - recommended |
|
60 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q6_K.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q6_k.gguf) | Q6_K | 6 | 5.94 GB| very large, extremely low quality loss |
|
61 |
+
| [Merak-7B-v4-PROTOTYPE6-model-Q8_0.gguf](https://huggingface.co/Ichsan2895/Merak-7B-v4-PROTOTYPE6-GGUF/blob/main/Merak-7B-v4-PROTOTYPE6-model-q8_0.gguf) | Q8_0 | 8 | 7.7 GB| very large, extremely low quality loss - not recommended |
|
62 |
|
63 |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
|
64 |
|