Update README.md
Browse files
README.md
CHANGED
|
@@ -18,12 +18,12 @@ The DeepSeek-R1 models include distilled and full-scale variants of both **Qwen*
|
|
| 18 |
|
| 19 |
| No | Variant | Branch | Cortex CLI command |
|
| 20 |
| -- | ---------------------------------------------------------------------------------------------- | ------- | ------------------------------------------ |
|
| 21 |
-
| 1 | [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/cortexso/deepseek-r1/tree/1.5b) | 1.5b | `cortex run
|
| 22 |
-
| 2 | [DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/cortexso/deepseek-r1/tree/7b) | 7b | `cortex run
|
| 23 |
-
| 3 | [DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/cortexso/deepseek-r1/tree/8b) | 8b | `cortex run
|
| 24 |
-
| 4 | [DeepSeek-R1-Distill-Qwen-14B](https://huggingface.co/cortexso/deepseek-r1/tree/14b) | 14b | `cortex run
|
| 25 |
-
| 5 | [DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/cortexso/deepseek-r1/tree/32b) | 32b | `cortex run
|
| 26 |
-
| 6 | [DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/cortexso/deepseek-r1/tree/70b) | 70b | `cortex run
|
| 27 |
|
| 28 |
Each branch contains a default quantized version:
|
| 29 |
- **Qwen-1.5B:** q4-km
|
|
@@ -38,8 +38,7 @@ Each branch contains a default quantized version:
|
|
| 38 |
1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)
|
| 39 |
2. Use in Jan model Hub:
|
| 40 |
```text
|
| 41 |
-
cortexso/deepseek-r1
|
| 42 |
-
cortexso/deepseek-r1 [WIP]
|
| 43 |
```
|
| 44 |
|
| 45 |
## Use it with Cortex (CLI)
|
|
@@ -47,11 +46,8 @@ Each branch contains a default quantized version:
|
|
| 47 |
1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)
|
| 48 |
2. Run the model with command:
|
| 49 |
```bash
|
| 50 |
-
cortex run
|
| 51 |
```
|
| 52 |
-
or
|
| 53 |
-
```bash
|
| 54 |
-
cortex run [WIP]
|
| 55 |
```
|
| 56 |
|
| 57 |
## Credits
|
|
|
|
| 18 |
|
| 19 |
| No | Variant | Branch | Cortex CLI command |
|
| 20 |
| -- | ---------------------------------------------------------------------------------------------- | ------- | ------------------------------------------ |
|
| 21 |
+
| 1 | [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/cortexso/deepseek-r1/tree/1.5b) | 1.5b | `cortex run deepseek-r1:1.5b` |
|
| 22 |
+
| 2 | [DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/cortexso/deepseek-r1/tree/7b) | 7b | `cortex run deepseek-r1:7b` |
|
| 23 |
+
| 3 | [DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/cortexso/deepseek-r1/tree/8b) | 8b | `cortex run deepseek-r1:8b` |
|
| 24 |
+
| 4 | [DeepSeek-R1-Distill-Qwen-14B](https://huggingface.co/cortexso/deepseek-r1/tree/14b) | 14b | `cortex run deepseek-r1:14b` |
|
| 25 |
+
| 5 | [DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/cortexso/deepseek-r1/tree/32b) | 32b | `cortex run deepseek-r1:32b` |
|
| 26 |
+
| 6 | [DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/cortexso/deepseek-r1/tree/70b) | 70b | `cortex run deepseek-r1:70b` |
|
| 27 |
|
| 28 |
Each branch contains a default quantized version:
|
| 29 |
- **Qwen-1.5B:** q4-km
|
|
|
|
| 38 |
1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)
|
| 39 |
2. Use in Jan model Hub:
|
| 40 |
```text
|
| 41 |
+
cortexso/deepseek-r1
|
|
|
|
| 42 |
```
|
| 43 |
|
| 44 |
## Use it with Cortex (CLI)
|
|
|
|
| 46 |
1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)
|
| 47 |
2. Run the model with command:
|
| 48 |
```bash
|
| 49 |
+
cortex run deepseek-r1
|
| 50 |
```
|
|
|
|
|
|
|
|
|
|
| 51 |
```
|
| 52 |
|
| 53 |
## Credits
|