Venice.aiThe combination of decentralized computing and AI agents creates powerful possibilities for developers. Venice's privacy-first API paired with Akash's decentralized compute marketplace and the ElizaOS agent framework allows you to deploy automated agents without sacrificing data privacy or creative freedom.
This guide walks you through launching an ElizaOS agent on Akash infrastructure using Venice API for inference, enabling you to create a character-based agent that can interact via API and post to social media. You are enot required to download or build any code packages, and it allows for simple “one-click” agent deployment with edits to a template to build your own character profile.
What You'll Need
Before getting started, make sure you have:
Venice API key (via Pro account, VVV staking or topping up your account with USD)
Akash account with a funded wallet ($AKT tokens)
Twitter/X account for the agent (optional but recommended)
Understanding the Components
ElizaOS Framework
ElizaOS is an open-source agent framework designed for building character-based AI agents with various capabilities. The framework supports multiple plugins including social media integrations, database storage, and more, making it ideal for creating persistent AI personalities.
Akash Network
Akash provides decentralized computing resources through a marketplace model. Users can deploy docker containers or VMs using pre-defined templates, making it perfect for launching AI agents without relying on centralized cloud providers.
Venice API
Venice's API provides access to powerful open-source AI models while maintaining complete privacy. Unlike other AI providers, Venice doesn't store your conversations or monitor your agents' activities, ensuring your deployment remains truly private.
Step-by-Step Deployment Guide of Your ElizaOS Agent Using Akash & Venice API
1. Go to Akash.network and click “Deploy Now” on the top right corner, which will bring you to console.akash.network
2. To use Akash, you will need a Keplr or Leap Cosmos wallet with AKT tokens on the Cosmos chain. We recommend you following the Akash guides to properly setup and fund those wallet. The simplest method of funding is to purchase AKT on Coinbase, and sending the funds directly to the kash wallet that has been setup. The simplest method of funding is to purchase AKT on Coinbase, and sending the funds directly to the akash wallet that has been setup
3. Once funded, you will see the USD value of your tokens on the top right corner within the wallet connection on the Akash console
4. Now that you are funded, you can click “Templates” to get started
5. Search for “Venice”, and click on the “Venice-ElizaOS” template
6. Next, click “Deploy”, which will open a new page to start configuring your Akash SDL
7. Now is time to configure your deployment. There are 3 key areas of this file that are important: A) Venice API Key, B) Twitter Configurations (only if you will be enabling twitter), C) Character Card InfoWhen properly configured, it will look something like this. Ensure all comments are removed prior to deployment.
Venice API Key: To generate an API key you can follow these instructions. You can get an API key through Venice with a pro account (explorer tier), by staking VVV to get ongoing access to daily compute (via VCUs), or by depositing USD (both will enable paid tier).
Twitter Configurations: This is where you enter your twitter username, password, and information about the frequency of posting from your character. It is important to note that Twitter integration is the most common failure of agent startup, and it may require authenticating account within twitter if it suspects suspicious activity. If it fails because of this, try logging in and out of twitter on your browser and re-deploying. You can choose to leave this section empty, and change to “CHAR_CLIENTS=” to disable the twitter integration.
Character Card Info: This defines the character, and how it interacts. This is the key initial information that structures the personality. Please refer to ElizaOS github for some examples of strong character cards. Note that for this SDL you must put everything on the same “line”, separated by commas. Do not use double quotes in these sections.
8. Click “Create Deployment”
9. Click continue to put 0.5 AKT into “escrow” of the deployment. You can add more funds later if deployed corrects.
10. Click OK/Sign when your connected wallet asks for the wallet signature
11. Choose a provider from the list that populates. We recommend only using “Audited” providers by Akash. You should also choose a provider within your Country to remove any additional possible risks for failure with your twitter integration (this is just a tip, not a requirement). Click “Accept Bid” to continue
12. You will be asked again to Approve a transaction in your associated wallet
13. When the deployment starts, you will see similar logs within the “Events” section of your deployment. This shows the successful download and startup of the docker image
14. Check the status of the Agent in the “Logs” section. This is a fairly heavy file, so it can take a few minutes for the agent to startup and begin seeing information in the logs section. You need to refresh often if you want to catch the beginning of the build in the actual UI. If you miss the startup messaging, you can also click “Download logs” to see exactly what happened during the build process. The image below shows that the build was successful
When the agent is running, you will see logs similar to this:
15. Next, click the “Leases” button to see the ports that the agent is hosting on, you will need to copy these down. Click the link to port 3000 within the “Forwarded Ports” section
16. If the agent is running and hosting properly, you will see this page in your web browser
17. Now you can interact with your character through API request. Note: There is also a web based chat UI, but it is still being debugged. We hope to have this fixed and updated soon. You can see the 5173 and 5174 posts were exposed in the SDL file for this purpose.
18. We recommend using postman to simplify the debugging of the API calls for testing purposes. The first thing you will need to do is send a “GET” call to get your Agent ID.
API Call:
curl --location 'http://<provider>:<port>/agents'
API Response:
{
"agents": [
{
"id": "<agentid>",
"name": "snoop",
"clients": [
"twitter"
]
}
]
}19. Now that you have your agent id, you can chat with the agent through a POST message.
API Call:
curl --location 'http://<provider>:<port>/<agentid>/message' \
--header 'Content-Type: application/json' \
--data '{
"text": "what is your name"
}'API Response:
[
{
"user": "snoop",
"text": "My name's Snoop, nephew, from Long Beach, fo shizzle. Ya dig?",
"action": "NONE"
}
]20. You will also see that your agent is posting properly to twitter. Check twitter to see what your agent is saying, and who it is interacting with. The agent will post and reply to tweets, and frequently interact with the accounts you set within the character card.
21. Congratulations on deploying your Eliza Agent using Venice API. Check back here for additional updates, as this is a living document.
Advanced Configurations
For more advanced deployments, you can customize:
Memory persistence with database integrations
Custom plugins for specialized functionality
Advanced prompt engineering for more nuanced responses
Additional social media integrations
Troubleshooting
Common issues and solutions:
Twitter Authentication Failures: Try logging into Twitter manually and checking for security prompts
API Connection Issues: Verify your Venice API key has the correct permissions
Deployment Errors: Check logs in the Akash console for detailed error messages
Character Definition Problems: Ensure your character definition follows the correct format without double quotes
Deploy your ElizaOS Agent with Akash & Venice API
By combining Venice's privacy-focused AI with Akash's decentralized compute and the ElizaOS agent framework, you've created an AI agent that maintains full privacy while offering powerful capabilities.
For more information and next steps:
Check the Venice API documentation
Review detailed model specifications in our model endpoints list
Join the Venice Discord for developer discussions and support
For updates and enhancements to this guide, check back regularly as we continue to improve this integration.
Back to all posts