Commit
•
1a13781
1
Parent(s):
cffaa8c
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -42,16 +42,27 @@ more details, including on how to concatenate multi-part files.
|
|
| 42 |
|
| 43 |
| Link | Type | Size/GB | Notes |
|
| 44 |
|:-----|:-----|--------:|:------|
|
|
|
|
| 45 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ1_M.gguf) | i1-IQ1_M | 5.0 | mostly desperate |
|
|
|
|
|
|
|
|
|
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ2_M.gguf) | i1-IQ2_M | 7.2 | |
|
| 47 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q2_K_S.gguf) | i1-Q2_K_S | 7.2 | very low quality |
|
| 48 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q2_K.gguf) | i1-Q2_K | 8.0 | IQ3_XXS probably better |
|
| 49 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 8.2 | lower quality |
|
|
|
|
|
|
|
|
|
|
| 50 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_M.gguf) | i1-IQ3_M | 9.7 | |
|
| 51 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q3_K_M.gguf) | i1-Q3_K_M | 10.7 | IQ3_S probably better |
|
| 52 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ4_XS.gguf) | i1-IQ4_XS | 11.0 | |
|
|
|
|
| 53 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q4_K_S.gguf) | i1-Q4_K_S | 11.8 | optimal size/speed/quality |
|
|
|
|
| 54 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q4_K_M.gguf) | i1-Q4_K_M | 12.9 | fast, recommended |
|
|
|
|
|
|
|
| 55 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q6_K.gguf) | i1-Q6_K | 16.7 | practically like static Q6_K |
|
| 56 |
|
| 57 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
|
|
|
| 42 |
|
| 43 |
| Link | Type | Size/GB | Notes |
|
| 44 |
|:-----|:-----|--------:|:------|
|
| 45 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ1_S.gguf) | i1-IQ1_S | 4.6 | for the desperate |
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ1_M.gguf) | i1-IQ1_M | 5.0 | mostly desperate |
|
| 47 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 5.7 | |
|
| 48 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ2_XS.gguf) | i1-IQ2_XS | 6.3 | |
|
| 49 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ2_S.gguf) | i1-IQ2_S | 6.6 | |
|
| 50 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ2_M.gguf) | i1-IQ2_M | 7.2 | |
|
| 51 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q2_K_S.gguf) | i1-Q2_K_S | 7.2 | very low quality |
|
| 52 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q2_K.gguf) | i1-Q2_K | 8.0 | IQ3_XXS probably better |
|
| 53 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 8.2 | lower quality |
|
| 54 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_XS.gguf) | i1-IQ3_XS | 8.8 | |
|
| 55 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_S.gguf) | i1-IQ3_S | 9.0 | beats Q3_K* |
|
| 56 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q3_K_S.gguf) | i1-Q3_K_S | 9.0 | IQ3_XS probably better |
|
| 57 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ3_M.gguf) | i1-IQ3_M | 9.7 | |
|
| 58 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q3_K_M.gguf) | i1-Q3_K_M | 10.7 | IQ3_S probably better |
|
| 59 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-IQ4_XS.gguf) | i1-IQ4_XS | 11.0 | |
|
| 60 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q4_0.gguf) | i1-Q4_0 | 11.7 | fast, low quality |
|
| 61 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q4_K_S.gguf) | i1-Q4_K_S | 11.8 | optimal size/speed/quality |
|
| 62 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q3_K_L.gguf) | i1-Q3_K_L | 11.8 | IQ3_M probably better |
|
| 63 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q4_K_M.gguf) | i1-Q4_K_M | 12.9 | fast, recommended |
|
| 64 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q5_K_S.gguf) | i1-Q5_K_S | 14.1 | |
|
| 65 |
+
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q5_K_M.gguf) | i1-Q5_K_M | 14.9 | |
|
| 66 |
| [GGUF](https://huggingface.co/mradermacher/granite-20b-code-instruct-r1.1-i1-GGUF/resolve/main/granite-20b-code-instruct-r1.1.i1-Q6_K.gguf) | i1-Q6_K | 16.7 | practically like static Q6_K |
|
| 67 |
|
| 68 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|