iAkashPaul
commited on
Commit
•
fc45f11
1
Parent(s):
c085c4b
Update README.md
Browse files
README.md
CHANGED
@@ -15,8 +15,11 @@ Contains Q4 & Q8 quantized GGUFs for [google/gemma](https://huggingface.co/colle
|
|
15 |
|
16 |
| Variant | Device | Perf |
|
17 |
| - | - | - |
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
| F16 | M1 Pro 10-core GPU | 30 tok/s|
|
19 |
-
|
|
20 |
-
| | M1 Pro 10-core GPU | 90 tok/s |
|
21 |
-
| Q8 | RTX 2070S | 25 tok/s |
|
22 |
-
| | M1 Pro 10-core GPU | 54 tok/s |
|
|
|
15 |
|
16 |
| Variant | Device | Perf |
|
17 |
| - | - | - |
|
18 |
+
| Q4 | M1 Pro 10-core GPU | 90 tok/s |
|
19 |
+
| | Snapdragon 778G CPU | 10 tok/s |
|
20 |
+
| | RTX 2070S | 40 tok/s |
|
21 |
+
| Q8 | M1 Pro 10-core GPU | 54 tok/s |
|
22 |
+
| | Snapdragon 778G CPU | 6 tok/s |
|
23 |
+
| | RTX 2070S | 25 tok/s |
|
24 |
| F16 | M1 Pro 10-core GPU | 30 tok/s|
|
25 |
+
| | Snapdragon 778G CPU | <1 tok/s |
|
|
|
|
|
|