auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -42,8 +42,20 @@ more details, including on how to concatenate multi-part files.
|
|
| 42 |
|
| 43 |
| Link | Type | Size/GB | Notes |
|
| 44 |
|:-----|:-----|--------:|:------|
|
|
|
|
| 45 |
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.8 | multi-modal supplement |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.mmproj-f16.gguf) | mmproj-f16 | 1.4 | multi-modal supplement |
|
|
|
|
|
|
|
| 47 |
|
| 48 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 49 |
types (lower is better):
|
|
|
|
| 42 |
|
| 43 |
| Link | Type | Size/GB | Notes |
|
| 44 |
|:-----|:-----|--------:|:------|
|
| 45 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q2_K.gguf) | Q2_K | 0.8 | |
|
| 46 |
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.8 | multi-modal supplement |
|
| 47 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q3_K_S.gguf) | Q3_K_S | 0.9 | |
|
| 48 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q3_K_M.gguf) | Q3_K_M | 0.9 | lower quality |
|
| 49 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q3_K_L.gguf) | Q3_K_L | 1.0 | |
|
| 50 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.IQ4_XS.gguf) | IQ4_XS | 1.0 | |
|
| 51 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q4_K_S.gguf) | Q4_K_S | 1.0 | fast, recommended |
|
| 52 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q4_K_M.gguf) | Q4_K_M | 1.1 | fast, recommended |
|
| 53 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q5_K_S.gguf) | Q5_K_S | 1.2 | |
|
| 54 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q5_K_M.gguf) | Q5_K_M | 1.2 | |
|
| 55 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q6_K.gguf) | Q6_K | 1.4 | very good quality |
|
| 56 |
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.mmproj-f16.gguf) | mmproj-f16 | 1.4 | multi-modal supplement |
|
| 57 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.Q8_0.gguf) | Q8_0 | 1.7 | fast, best quality |
|
| 58 |
+
| [GGUF](https://huggingface.co/mradermacher/UME-R1-2B-GGUF/resolve/main/UME-R1-2B.f16.gguf) | f16 | 3.2 | 16 bpw, overkill |
|
| 59 |
|
| 60 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 61 |
types (lower is better):
|