Upload complete model
Browse files
README.md
CHANGED
@@ -6,11 +6,11 @@ tags:
|
|
6 |
- mlx
|
7 |
pipeline_tag: text-generation
|
8 |
---
|
9 |
-
## CURRENTLY UPLOADING FILES
|
10 |
This notice will be removed once all files have been uploaded.
|
11 |
-
...
|
12 |
|
13 |
-
|
|
|
14 |
|
15 |
*q5.5bit quant typically achieves 1.141 perplexity in our testing*
|
16 |
| Quantization | Perplexity |
|
@@ -26,6 +26,6 @@ This notice will be removed once all files have been uploaded.
|
|
26 |
|
27 |
* Runs on a single M3 Ultra 512GB RAM
|
28 |
* Memory usage: ~480 GB
|
29 |
-
* Expect ~
|
30 |
* Built with a modified version of [MLX](https://github.com/ml-explore/mlx) 0.26
|
31 |
-
* For more details see [demonstration video](https://
|
|
|
6 |
- mlx
|
7 |
pipeline_tag: text-generation
|
8 |
---
|
9 |
+
## ----CURRENTLY UPLOADING FILES-----
|
10 |
This notice will be removed once all files have been uploaded.
|
|
|
11 |
|
12 |
+
## Notes
|
13 |
+
**See DeepSeek-V3.1 5.5bit MLX in action - [demonstration video](hhttps://youtu.be/ufXZI6aqOU8)**
|
14 |
|
15 |
*q5.5bit quant typically achieves 1.141 perplexity in our testing*
|
16 |
| Quantization | Perplexity |
|
|
|
26 |
|
27 |
* Runs on a single M3 Ultra 512GB RAM
|
28 |
* Memory usage: ~480 GB
|
29 |
+
* Expect ~13-19 tokens/s
|
30 |
* Built with a modified version of [MLX](https://github.com/ml-explore/mlx) 0.26
|
31 |
+
* For more details see [demonstration video](https://youtu.be/ufXZI6aqOU8) or visit [DeepSeek-V3.1](https://huggingface.co/deepseek-ai/DeepSeek-V3.1).
|