Upload complete model
Browse files
README.md
CHANGED
|
@@ -6,9 +6,12 @@ tags:
|
|
| 6 |
- mlx
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
|
|
|
|
|
|
|
|
|
| 9 |
**See DeepSeek-V3.1 5.5bit MLX in action - [demonstration video](https://youtube.com/xcreate)**
|
| 10 |
|
| 11 |
-
*q5.5bit quant typically achieves 1.141 perplexity in our testing
|
| 12 |
| Quantization | Perplexity |
|
| 13 |
|:------------:|:----------:|
|
| 14 |
| **q2.5** | 41.293 |
|
|
@@ -20,7 +23,8 @@ pipeline_tag: text-generation
|
|
| 20 |
|
| 21 |
## Usage Notes
|
| 22 |
|
| 23 |
-
*
|
| 24 |
* Memory usage: ~480 GB
|
| 25 |
* Expect ~15 tokens/s
|
|
|
|
| 26 |
* For more details see [demonstration video](https://youtube.com/xcreate) or visit [DeepSeek-V3.1](https://huggingface.co/deepseek-ai/DeepSeek-V3.1).
|
|
|
|
| 6 |
- mlx
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
| 9 |
+
## CURRENTLY UPLOADING FILES
|
| 10 |
+
This notice will be removed once all files are uploaded.
|
| 11 |
+
|
| 12 |
**See DeepSeek-V3.1 5.5bit MLX in action - [demonstration video](https://youtube.com/xcreate)**
|
| 13 |
|
| 14 |
+
*q5.5bit quant typically achieves 1.141 perplexity in our testing*
|
| 15 |
| Quantization | Perplexity |
|
| 16 |
|:------------:|:----------:|
|
| 17 |
| **q2.5** | 41.293 |
|
|
|
|
| 23 |
|
| 24 |
## Usage Notes
|
| 25 |
|
| 26 |
+
* Runs on a single M3 Ultra 512GB RAM
|
| 27 |
* Memory usage: ~480 GB
|
| 28 |
* Expect ~15 tokens/s
|
| 29 |
+
* Built with a modified version of [MLX](https://github.com/ml-explore/mlx) 0.26
|
| 30 |
* For more details see [demonstration video](https://youtube.com/xcreate) or visit [DeepSeek-V3.1](https://huggingface.co/deepseek-ai/DeepSeek-V3.1).
|