Update README.md
Browse files
README.md
CHANGED
|
@@ -5,6 +5,21 @@ license: gpl-3.0
|
|
| 5 |
# NVOM.ai LLama 3.2 3b
|
| 6 |
A lightweight model optimized and trained by nvom.ai
|
| 7 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
# Run model locally
|
| 9 |
You can use [LM STUDIO](https://lmstudio.ai/) to start model
|
| 10 |
Or you can use code and Transformers
|
|
|
|
| 5 |
# NVOM.ai LLama 3.2 3b
|
| 6 |
A lightweight model optimized and trained by nvom.ai
|
| 7 |
|
| 8 |
+
Benchmarks:
|
| 9 |
+
|
| 10 |
+
NVOM.ai AIBench SmartBench Specials (Based on NVOM.ai bench)
|
| 11 |
+
NVOM Preview 4b 7.91 10.12 9.49 Speed, Smart, Quanatisation
|
| 12 |
+
NVOM LLama 3.2 3b 6.11 4.95 3.31 Speed, Quanatisation
|
| 13 |
+
Meta LLama 3.2 3b 3.76 2.11 2.93 Quanatisation
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
1-4 - bad
|
| 18 |
+
|
| 19 |
+
5-8 - normal
|
| 20 |
+
|
| 21 |
+
8-12 - best
|
| 22 |
+
|
| 23 |
# Run model locally
|
| 24 |
You can use [LM STUDIO](https://lmstudio.ai/) to start model
|
| 25 |
Or you can use code and Transformers
|