auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -22,7 +22,7 @@ quantized_by: mradermacher
|
|
| 22 |
weighted/imatrix quants of https://huggingface.co/google/codegemma-7b
|
| 23 |
|
| 24 |
<!-- provided-files -->
|
| 25 |
-
static quants are available at https://huggingface.co/mradermacher/
|
| 26 |
## Usage
|
| 27 |
|
| 28 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
@@ -35,31 +35,31 @@ more details, including on how to concatenate multi-part files.
|
|
| 35 |
|
| 36 |
| Link | Type | Size/GB | Notes |
|
| 37 |
|:-----|:-----|--------:|:------|
|
| 38 |
-
| [GGUF](https://huggingface.co/mradermacher/
|
| 39 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 40 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 41 |
-
| [GGUF](https://huggingface.co/mradermacher/
|
| 42 |
-
| [GGUF](https://huggingface.co/mradermacher/
|
| 43 |
-
| [GGUF](https://huggingface.co/mradermacher/
|
| 44 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 45 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 46 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 47 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 48 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 49 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 50 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 51 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 52 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 53 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 54 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 55 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 56 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 57 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 58 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 59 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 60 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 61 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 62 |
-
| [PART 1](https://huggingface.co/mradermacher/
|
| 63 |
|
| 64 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 65 |
types (lower is better):
|
|
|
|
| 22 |
weighted/imatrix quants of https://huggingface.co/google/codegemma-7b
|
| 23 |
|
| 24 |
<!-- provided-files -->
|
| 25 |
+
static quants are available at https://huggingface.co/mradermacher/codegemma-7b-GGUF
|
| 26 |
## Usage
|
| 27 |
|
| 28 |
If you are unsure how to use GGUF files, refer to one of [TheBloke's
|
|
|
|
| 35 |
|
| 36 |
| Link | Type | Size/GB | Notes |
|
| 37 |
|:-----|:-----|--------:|:------|
|
| 38 |
+
| [GGUF](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.3 | very low quality |
|
| 39 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ1_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ1_S.gguf) | i1-IQ1_S | 4.4 | for the desperate |
|
| 40 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ1_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ1_M.gguf) | i1-IQ1_M | 4.7 | mostly desperate |
|
| 41 |
+
| [GGUF](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 5.1 | fast on arm, low quality |
|
| 42 |
+
| [GGUF](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 5.1 | fast on arm+i8mm, low quality |
|
| 43 |
+
| [GGUF](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 5.1 | fast on arm+sve, low quality |
|
| 44 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ2_XXS.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 5.3 | |
|
| 45 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ2_XS.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 5.7 | |
|
| 46 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ2_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ2_S.gguf) | i1-IQ2_S | 5.9 | |
|
| 47 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ2_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ2_M.gguf) | i1-IQ2_M | 6.4 | |
|
| 48 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q2_K.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q2_K.gguf) | i1-Q2_K | 7.1 | IQ3_XXS probably better |
|
| 49 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ3_XXS.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 7.1 | lower quality |
|
| 50 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ3_XS.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ3_XS.gguf) | i1-IQ3_XS | 7.7 | |
|
| 51 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ3_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ3_S.gguf) | i1-IQ3_S | 8.1 | beats Q3_K* |
|
| 52 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q3_K_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q3_K_S.gguf) | i1-Q3_K_S | 8.1 | IQ3_XS probably better |
|
| 53 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ3_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ3_M.gguf) | i1-IQ3_M | 8.3 | |
|
| 54 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q3_K_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q3_K_M.gguf) | i1-Q3_K_M | 8.8 | IQ3_S probably better |
|
| 55 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q3_K_L.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q3_K_L.gguf) | i1-Q3_K_L | 9.5 | IQ3_M probably better |
|
| 56 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-IQ4_XS.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-IQ4_XS.gguf) | i1-IQ4_XS | 9.6 | |
|
| 57 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q4_0.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_0.gguf) | i1-Q4_0 | 10.2 | fast, low quality |
|
| 58 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q4_K_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_K_S.gguf) | i1-Q4_K_S | 10.2 | optimal size/speed/quality |
|
| 59 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q4_K_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q4_K_M.gguf) | i1-Q4_K_M | 10.8 | fast, recommended |
|
| 60 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q5_K_S.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q5_K_S.gguf) | i1-Q5_K_S | 12.1 | |
|
| 61 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q5_K_M.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q5_K_M.gguf) | i1-Q5_K_M | 12.4 | |
|
| 62 |
+
| [PART 1](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/CodeGemma-7b.i1-Q6_K.gguf) [PART 2](https://huggingface.co/mradermacher/codegemma-7b-i1-GGUF/resolve/main/codegemma-7b.i1-Q6_K.gguf) | i1-Q6_K | 14.1 | practically like static Q6_K |
|
| 63 |
|
| 64 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 65 |
types (lower is better):
|