feat: add frps/frpc/mineru-sgalng
This commit is contained in:
26
src/mineru-sgalng/README.md
Normal file
26
src/mineru-sgalng/README.md
Normal file
@@ -0,0 +1,26 @@
|
||||
# MinerU SGLang
|
||||
|
||||
SGLang backend server:
|
||||
|
||||
```bash
|
||||
docker compose --profile sglang-server up -d
|
||||
```
|
||||
|
||||
Document parse API:
|
||||
|
||||
```bash
|
||||
docker compose --profile api up -d
|
||||
```
|
||||
|
||||
Gradio WebUI:
|
||||
|
||||
```bash
|
||||
docker compose --profile gradio up -d
|
||||
```
|
||||
|
||||
Test SGLang backend:
|
||||
|
||||
```bash
|
||||
pip install mineru
|
||||
mineru -p demo.pdf -o ./output -b vlm-sglang-client -u http://localhost:30000
|
||||
```
|
||||
90
src/mineru-sgalng/docker-compose.yaml
Normal file
90
src/mineru-sgalng/docker-compose.yaml
Normal file
@@ -0,0 +1,90 @@
|
||||
x-default: &default
|
||||
restart: unless-stopped
|
||||
volumes:
|
||||
- &localtime /etc/localtime:/etc/localtime:ro
|
||||
- &timezone /etc/timezone:/etc/timezone:ro
|
||||
logging:
|
||||
driver: json-file
|
||||
options:
|
||||
max-size: 100m
|
||||
|
||||
x-mineru-sglang: &mineru-sglang
|
||||
<<: *default
|
||||
image: ${MINERU_DOCKER_IMAGE:-alexsuntop/mineru-sglang:2.2.2}
|
||||
environment:
|
||||
MINERU_MODEL_SOURCE: local
|
||||
ulimits:
|
||||
memlock: -1
|
||||
stack: 67108864
|
||||
ipc: host
|
||||
deploy:
|
||||
resources:
|
||||
limits:
|
||||
cpus: '8.0'
|
||||
memory: 4G
|
||||
reservations:
|
||||
cpus: '1.0'
|
||||
memory: 2G
|
||||
devices:
|
||||
- driver: nvidia
|
||||
device_ids: [ '0' ]
|
||||
capabilities: [ gpu ]
|
||||
|
||||
services:
|
||||
mineru-sglang-server:
|
||||
<<: *mineru-sglang
|
||||
profiles: ["sglang-server"]
|
||||
ports:
|
||||
- ${MINERU_PORT_OVERRIDE_SGLANG:-30000}:30000
|
||||
entrypoint: mineru-sglang-server
|
||||
command:
|
||||
- --host 0.0.0.0
|
||||
- --port 30000
|
||||
|
||||
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
|
||||
# - --data-parallel-size 2
|
||||
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
|
||||
# if VRAM issues persist, try lowering it further to `0.4` or below.
|
||||
# - --gpu-memory-utilization 0.5
|
||||
|
||||
healthcheck:
|
||||
test: ["CMD-SHELL", "curl -f http://localhost:30000/health || exit 1"]
|
||||
|
||||
mineru-api:
|
||||
<<: *mineru-sglang
|
||||
profiles: ["api"]
|
||||
ports:
|
||||
- ${MINERU_PORT_OVERRIDE_API:-8000}:8000
|
||||
entrypoint: mineru-api
|
||||
command:
|
||||
- --host 0.0.0.0
|
||||
- --port 8000
|
||||
|
||||
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
|
||||
# - --data-parallel-size 2
|
||||
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
|
||||
# if VRAM issues persist, try lowering it further to `0.4` or below.
|
||||
# - --gpu-memory-utilization 0.5
|
||||
|
||||
mineru-gradio:
|
||||
<<: *mineru-sglang
|
||||
profiles: ["gradio"]
|
||||
ports:
|
||||
- ${MINERU_PORT_OVERRIDE_GRADIO:-7860}:7860
|
||||
entrypoint: mineru-gradio
|
||||
command:
|
||||
- --server-name 0.0.0.0
|
||||
- --server-port 7860
|
||||
|
||||
# Enable the vllm engine for Gradio
|
||||
- --enable-vllm-engine true
|
||||
# If you want to disable the API, set this to false
|
||||
# - --enable-api false
|
||||
# If you want to limit the number of pages for conversion, set this to a specific number
|
||||
# - --max-convert-pages 20
|
||||
|
||||
# If using multiple GPUs, increase throughput using vllm's multi-GPU parallel mode.
|
||||
# - --data-parallel-size 2
|
||||
# If running on a single GPU and encountering VRAM shortage, reduce the KV cache size by this parameter,
|
||||
# if VRAM issues persist, try lowering it further to `0.4` or below.
|
||||
# - --gpu-memory-utilization 0.5
|
||||
Reference in New Issue
Block a user