Commit
•
dd9d92f
1
Parent(s):
22ad41f
auto-patch README.md
Browse files
README.md
CHANGED
|
@@ -27,7 +27,21 @@ more details, including on how to concatenate multi-part files.
|
|
| 27 |
|
| 28 |
| Link | Type | Size/GB | Notes |
|
| 29 |
|:-----|:-----|--------:|:------|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q8_0.gguf) | Q8_0 | 7.3 | fast, best quality |
|
|
|
|
| 31 |
|
| 32 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 33 |
types (lower is better):
|
|
|
|
| 27 |
|
| 28 |
| Link | Type | Size/GB | Notes |
|
| 29 |
|:-----|:-----|--------:|:------|
|
| 30 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q2_K.gguf) | Q2_K | 2.6 | |
|
| 31 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.IQ3_XS.gguf) | IQ3_XS | 2.9 | |
|
| 32 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.IQ3_S.gguf) | IQ3_S | 3.0 | beats Q3_K* |
|
| 33 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q3_K_S.gguf) | Q3_K_S | 3.0 | |
|
| 34 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.IQ3_M.gguf) | IQ3_M | 3.2 | |
|
| 35 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q3_K_M.gguf) | Q3_K_M | 3.4 | lower quality |
|
| 36 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q3_K_L.gguf) | Q3_K_L | 3.7 | |
|
| 37 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.IQ4_XS.gguf) | IQ4_XS | 3.7 | |
|
| 38 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q4_K_S.gguf) | Q4_K_S | 4.0 | fast, recommended |
|
| 39 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q4_K_M.gguf) | Q4_K_M | 4.2 | fast, recommended |
|
| 40 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q5_K_S.gguf) | Q5_K_S | 4.8 | |
|
| 41 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q5_K_M.gguf) | Q5_K_M | 4.9 | |
|
| 42 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q6_K.gguf) | Q6_K | 5.6 | very good quality |
|
| 43 |
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.Q8_0.gguf) | Q8_0 | 7.3 | fast, best quality |
|
| 44 |
+
| [GGUF](https://huggingface.co/mradermacher/MuMath-Code-L-7B-GGUF/resolve/main/MuMath-Code-L-7B.f16.gguf) | f16 | 13.6 | 16 bpw, overkill |
|
| 45 |
|
| 46 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
| 47 |
types (lower is better):
|