Update README.md
Browse files
README.md
CHANGED
|
@@ -6,6 +6,7 @@ tags:
|
|
| 6 |
- llama.cpp
|
| 7 |
- quantized
|
| 8 |
- iq4_xs
|
|
|
|
| 9 |
- q5_k_m
|
| 10 |
- q6_k
|
| 11 |
- imatrix
|
|
@@ -13,7 +14,7 @@ tags:
|
|
| 13 |
- roleplay
|
| 14 |
- storytelling
|
| 15 |
- language-model
|
| 16 |
-
quantization: IQ4_XS, Q5_K_M, Q6_K
|
| 17 |
model_type: llama
|
| 18 |
language:
|
| 19 |
- en
|
|
@@ -27,8 +28,8 @@ This is a **quantized GGUF version** of [`saturated-labs/T-Rex-mini`](https://hu
|
|
| 27 |
|
| 28 |
- **Original Model**: `saturated-labs/T-Rex-mini`
|
| 29 |
- **Format**: GGUF (`.gguf`)
|
| 30 |
-
- **Quantization Type**: `IQ4_XS, Q5_K_M, Q6_K`
|
| 31 |
- **Tool Used**: [`llama.cpp`](https://github.com/ggerganov/llama.cpp)
|
| 32 |
- **Command**:
|
| 33 |
```bash
|
| 34 |
-
./llama-quantize.exe --imatrix imatrix.dat
|
|
|
|
| 6 |
- llama.cpp
|
| 7 |
- quantized
|
| 8 |
- iq4_xs
|
| 9 |
+
- q4_k_m
|
| 10 |
- q5_k_m
|
| 11 |
- q6_k
|
| 12 |
- imatrix
|
|
|
|
| 14 |
- roleplay
|
| 15 |
- storytelling
|
| 16 |
- language-model
|
| 17 |
+
quantization: IQ4_XS, Q4_K_M, Q5_K_M, Q6_K
|
| 18 |
model_type: llama
|
| 19 |
language:
|
| 20 |
- en
|
|
|
|
| 28 |
|
| 29 |
- **Original Model**: `saturated-labs/T-Rex-mini`
|
| 30 |
- **Format**: GGUF (`.gguf`)
|
| 31 |
+
- **Quantization Type**: `IQ4_XS, Q4_K_M, Q5_K_M, Q6_K`
|
| 32 |
- **Tool Used**: [`llama.cpp`](https://github.com/ggerganov/llama.cpp)
|
| 33 |
- **Command**:
|
| 34 |
```bash
|
| 35 |
+
./llama-quantize.exe --imatrix imatrix.dat t-rex-mini-f16.gguf t-rex-mini-QX_X_X.gguf QX_X_X
|