EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Doc] Fix DeepSeek-V3.2 tutorial. (#5190)

Browse Source 
### What this PR does / why we need it?
Fix DeepSeek-V3.2 tutorial.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

Signed-off-by: menogrey <1299267905@qq.com>
This commit is contained in:
zhangyiming
2025-12-22 11:30:17 +08:00
committed by GitHub
parent 492173cf89
commit dc047489c7
2 changed files with 20 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
docs/source/tutorials/DeepSeek-V3.2.md
View File

@@ -18,7 +18,7 @@ Refer to [feature guide](../user_guide/feature_guide/index.md) to get the featur
- `DeepSeek-V3.2-Exp`(BF16 version): require 2 Atlas 800 A3 (64G × 16) nodes or 4 Atlas 800 A2 (64G × 8) nodes. [Download model weight](https://modelers.cn/models/Modelers_Park/DeepSeek-V3.2-Exp-BF16)
- `DeepSeek-V3.2-Exp-w8a8`(Quantized version): require 1 Atlas 800 A3 (64G × 16) node or 2 Atlas 800 A2 (64G × 8) nodes. [Download model weight](https://modelers.cn/models/Modelers_Park/DeepSeek-V3.2-Exp-w8a8)
- `DeepSeek-V3.2`(BF16 version): require 2 Atlas 800 A3 (64G × 16) nodes or 4 Atlas 800 A2 (64G × 8) nodes. [Download model weight](https://modelscope.cn/models/deepseek-ai/DeepSeek-V3.2/)
- `DeepSeek-V3.2`(BF16 version): require 2 Atlas 800 A3 (64G × 16) nodes or 4 Atlas 800 A2 (64G × 8) nodes. Model weight in BF16 not found now.
- `DeepSeek-V3.2-w8a8`(Quantized version): require 1 Atlas 800 A3 (64G × 16) node or 2 Atlas 800 A2 (64G × 8) nodes. [Download model weight](https://modelers.cn/models/Eco-Tech/DeepSeek-V3.2-w8a8-QuaRot)
It is recommended to download the model weight to the shared directory of multiple nodes, such as `/root/.cache/`
@@ -29,10 +29,26 @@ If you want to deploy multi-node environment, you need to verify multi-node comm
### Installation
You can using our official docker image and install extra operator for supporting `DeepSeek-V3.2`.
You can using our official docker image to run `DeepSeek-V3.2` directly..
:::{note}
We strongly recommend you to install triton ascend package to speed up the inference.
The [Triton Ascend](https://gitee.com/ascend/triton-ascend) is for better performance, please follow the instructions below to install it and its dependency.
Source the Ascend BiSheng toolkit, execute the command:
```bash
source /usr/local/Ascend/ascend-toolkit/8.3.RC2/bisheng_toolkit/set_env.sh
```
Install Triton Ascend:
```bash
wget https://vllm-ascend.obs.cn-north-4.myhuaweicloud.com/vllm-ascend/triton_ascend-3.2.0.dev2025110717-cp311-cp311-manylinux_2_27_aarch64.whl
pip install triton_ascend-3.2.0.dev2025110717-cp311-cp311-manylinux_2_27_aarch64.whl
```
:::
:::::{tab-set}
@@ -638,7 +654,7 @@ Take the `serve` as an example. Run the code as follows.
```shell
export VLLM_USE_MODELSCOPE=true
vllm bench serve --model vllm-ascend/DeepSeek-V3.2-W8A8 --dataset-name random --random-input 200 --num-prompt 200 --request-rate 1 --save-result --result-dir ./
vllm bench serve --model /root/.cache/Eco-Tech/DeepSeek-V3.2-w8a8-QuaRot --dataset-name random --random-input 200 --num-prompt 200 --request-rate 1 --save-result --result-dir ./
```
After about several minutes, you can get the performance evaluation result. With this tutorial, the performance result is: