apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2026-04-23 00:17:25 +08:00
Code Issues Actions 19 Packages Projects Releases Wiki Activity

[Docs] PaddleOCR-VL add RTX3060 server param (#4765)

Browse Source 
* [Docs] PaddleOCR-VL add RTX3060 server param

* modify config

* fix bug
This commit is contained in:
ming1753
2025-11-03 19:55:05 +08:00
committed by GitHub
parent 8690cf8569
commit 35a6969a44
2 changed files with 35 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/zh/best_practices/PaddleOCR-VL-0.9B.md
+17 -4
View File
@@ -5,7 +5,7 @@
## 一、环境准备
### 1.1 支持情况
推荐硬件配置：
- 显存：24GB显存及以上
- 显存：12GB显存及以上
- 共享内存：2G及以上
### 1.2 安装fastdeploy
@@ -14,7 +14,20 @@
## 二、如何使用
### 2.1 基础：启动服务
**示例1：** 4090上单卡部署16K上下文的服务
**示例1：** 3060上单卡部署16K上下文的服务
```shell
python -m fastdeploy.entrypoints.openai.api_server \
--model PaddlePaddle/PaddleOCR-VL \
--port 8180 \
--metrics-port 8181 \
--engine-worker-queue-port 8182 \
--max-model-len 16384 \
--max-num-batched-tokens 16384 \
--gpu-memory-utilization 0.9 \
--max-num-seqs 128
```
**示例2：** 4090上单卡部署16K上下文的服务
```shell
python -m fastdeploy.entrypoints.openai.api_server \
--model PaddlePaddle/PaddleOCR-VL \
@@ -24,10 +37,10 @@ python -m fastdeploy.entrypoints.openai.api_server \
--max-model-len 16384 \
--max-num-batched-tokens 16384 \
--gpu-memory-utilization 0.8 \
--max-num-seqs 128
--max-num-seqs 196
```
**示例2：** A100上单卡部署16K上下文的服务
**示例3：** A100上单卡部署16K上下文的服务
```shell
python -m fastdeploy.entrypoints.openai.api_server \
--model PaddlePaddle/PaddleOCR-VL \