mradermacher
commited on
Commit
•
3ba4862
1
Parent(s):
a2087bd
auto-patch README.md
Browse files
README.md
CHANGED
@@ -40,7 +40,16 @@ more details, including on how to concatenate multi-part files.
|
|
40 |
| Link | Type | Size/GB | Notes |
|
41 |
|:-----|:-----|--------:|:------|
|
42 |
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q2_K.gguf) | Q2_K | 8.9 | |
|
|
|
|
|
|
|
|
|
43 |
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q4_K_S.gguf) | Q4_K_S | 13.8 | fast, recommended |
|
|
|
|
|
|
|
|
|
|
|
44 |
|
45 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
46 |
types (lower is better):
|
|
|
40 |
| Link | Type | Size/GB | Notes |
|
41 |
|:-----|:-----|--------:|:------|
|
42 |
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q2_K.gguf) | Q2_K | 8.9 | |
|
43 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q3_K_S.gguf) | Q3_K_S | 10.5 | |
|
44 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q3_K_M.gguf) | Q3_K_M | 11.7 | lower quality |
|
45 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q3_K_L.gguf) | Q3_K_L | 12.6 | |
|
46 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.IQ4_XS.gguf) | IQ4_XS | 13.1 | |
|
47 |
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q4_K_S.gguf) | Q4_K_S | 13.8 | fast, recommended |
|
48 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q4_K_M.gguf) | Q4_K_M | 14.7 | fast, recommended |
|
49 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q5_K_S.gguf) | Q5_K_S | 16.7 | |
|
50 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q5_K_M.gguf) | Q5_K_M | 17.2 | |
|
51 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q6_K.gguf) | Q6_K | 19.9 | very good quality |
|
52 |
+
| [GGUF](https://huggingface.co/mradermacher/Moe-4x7b-reason-code-qa-GGUF/resolve/main/Moe-4x7b-reason-code-qa.Q8_0.gguf) | Q8_0 | 25.8 | fast, best quality |
|
53 |
|
54 |
Here is a handy graph by ikawrakow comparing some lower-quality quant
|
55 |
types (lower is better):
|