Start your AI Gateway
The AI Subnet is not yet integrated into the main
go-livepeer software due to its
Alpha status. To enable AI inference capabilities on your Gateway node,
please use the ai-video
branch of
go-livepeer. This
branch contains the necessary software for the AI Gateway node. Currently, there
are two methods to run the AI Subnet software:
- Docker: This is the most straightforward and recommended method to run the AI Gateway node.
- Pre-built Binaries: If you prefer not to use Docker, pre-built binaries are available.
Start the AI Gateway
Please follow the steps below to start your AI Subnet Gateway node:
Retrieve the AI Subnet Docker Image
Fetch the latest AI Subnet Docker image from the Livepeer Docker Hub with the following command:
docker pull livepeer/go-livepeer:ai-video
Launch an (Offchain) AI Gateway
Execute the AI Subnet Docker image using the following command:
docker run \
--name livepeer_ai_gateway \
-v ~/.lpData2/:/root/.lpData2 \
-p 8937:8937 \
--network host \
livepeer/go-livepeer:ai-video \
-datadir ~/.lpData2 \
-gateway \
-orchAddr <ORCH_LIST> \
-httpAddr 0.0.0.0:8937 \
-v 6 \
-httpIngest
This launches an offchain AI Gateway node. The flags are similar to those used for a Mainnet Transcoding Network Gateway node. For more information, see the go-livepeer CLI reference.
Confirm Successful Startup of the AI Gateway
If your AI Subnet Gateway node is functioning correctly, you should see the following output:
I0501 11:07:47.609839 1 mediaserver.go:201] Transcode Job Type: [{P240p30fps16x9 600k 30 0 426x240 16:9 0 0 0s 0 0 0 0} {P360p30fps16x9 1200k 30 0 640x360 16:9 0 0 0s 0 0 0 0}]
I0501 11:07:47.609917 1 mediaserver.go:226] HTTP Server listening on http://0.0.0.0:8937
I0501 11:07:47.609963 1 lpms.go:92] LPMS Server listening on rtmp://127.0.0.1:1935
Check Port Availability
To make your AI Subnet Gateway node accessible from the internet, you need to configure your network settings. Ensure that port 8937
is unblocked on your machine. Additionally, consider setting up port forwarding on your router, which will allow the Gateway node to be reachable from the internet.
Confirm the AI Gateway is Operational
After launching the AI Subnet Gateway node, verify its operation by sending an AI inference request directly to it. This requires an active off-chain AI Orchestrator node. For guidance on setting up an AI Orchestrator node, refer to the AI Orchestrator Setup Guide.
Launch an AI Orchestrator
Start an AI Orchestrator node on port 8936
following the AI Orchestrator Setup Guide. Ensure it has loaded the desired model for inference (e.g., “ByteDance/SDXL-Lightning”).
Link Gateway to AI Orchestrator
To connect your Gateway node to the AI Orchestrator node, specify the Orchestrator’s address when launching the Gateway node. Replace <ORCH_LIST>
with the Orchestrator’s address, like so:
-orchAddr 0.0.0.0:8936
Submit an Inference Request
Refer to the AI API reference to understand how to submit an inference request to the Gateway node. For instance, to generate an image from text, use the following curl
command:
curl -X POST "http://0.0.0.0:8937/text-to-image" \
-H "Content-Type: application/json" \
-d '{
"model_id":"ByteDance/SDXL-Lightning",
"prompt":"A cool cat on the beach",
"width": 1024,
"height": 1024
}'
Inspect the Response
If the Gateway node is functioning correctly, you should receive a response similar to the following:
{
"images": [
{
"seed": 2562822894,
"url": "https://0.0.0.0:8937/stream/d0fc1fc6/8fdf5a94.png"
}
]
}
Consult the Text-to-image Pipeline Documentation for more details on interpreting the response.
Was this page helpful?