Signed-off-by: Ye, Xinyu <xinyu.ye@intel.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
22 lines
739 B
Markdown
22 lines
739 B
Markdown
# Text-to-Image Microservice
|
|
|
|
Text-to-Image is a task that generate image conditioning on the provided text. This microservice supports text-to-image task by using Stable Diffusion (SD) model.
|
|
|
|
## Deploy Text-to-Image Service
|
|
|
|
### Deploy Text-to-Image Service on Xeon
|
|
|
|
Refer to the [Xeon Guide](./docker_compose/intel/cpu/xeon/README.md) for detail.
|
|
|
|
### Deploy Text-to-Image Service on Gaudi
|
|
|
|
Refer to the [Gaudi Guide](./docker_compose/intel/hpu/gaudi/README.md) for detail.
|
|
|
|
## Consume Text-to-Image Service
|
|
|
|
Use below command to generate image.
|
|
|
|
```bash
|
|
http_proxy="" curl http://localhost:9379/v1/text2image -XPOST -d '{"prompt":"An astronaut riding a green horse", "num_images_per_prompt":1}' -H 'Content-Type: application/json'
|
|
```
|