Update README.md
Browse files
README.md
CHANGED
|
@@ -17,6 +17,12 @@ base_model: google/gemma-3-27b-it-qat-q4_0-unquantized
|
|
| 17 |
|
| 18 |
## Llamacpp imatrix Quantizations of gemma-3-27b-it-qat by google
|
| 19 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b5147">b5147</a> for quantization.
|
| 21 |
|
| 22 |
Original model: https://huggingface.co/google/gemma-3-27b-it-qat-q4_0-unquantized
|
|
|
|
| 17 |
|
| 18 |
## Llamacpp imatrix Quantizations of gemma-3-27b-it-qat by google
|
| 19 |
|
| 20 |
+
These are derived from the QAT (quantized aware training) weights provided by Google
|
| 21 |
+
|
| 22 |
+
*ONLY* Q4_0 is expected to be better, but figured while I'm at it I might as well make others to see what happens?
|
| 23 |
+
|
| 24 |
+
[gemma-3-27b-it-qat-Q4_0.gguf](https://huggingface.co/bartowski/google_gemma-3-27b-it-qat-GGUF/blob/main/google_gemma-3-27b-it-qat-Q4_0.gguf) | Q4_0 | 15.62GB | Should be improved due to QAT, offers online repacking for ARM and AVX CPU inference.
|
| 25 |
+
|
| 26 |
Using <a href="https://github.com/ggerganov/llama.cpp/">llama.cpp</a> release <a href="https://github.com/ggerganov/llama.cpp/releases/tag/b5147">b5147</a> for quantization.
|
| 27 |
|
| 28 |
Original model: https://huggingface.co/google/gemma-3-27b-it-qat-q4_0-unquantized
|