Venice sells subscription tiers (including the Diem tier) that include API access. If you have spare headroom in your Venice subscription, you can resell that capacity through Morpheus by configuring your proxy-router to forward to Venice’s API.
Read Venice’s TOS. Resale terms vary by tier and over time — confirm yours allows reselling API access before going live. This page describes mechanics, not legal permission.

​
Architecture

​
Steps

1

Get a Venice API key

Sign up at https://venice.ai and obtain an API key from your account dashboard. Confirm your tier supports the model you intend to resell at the concurrency you intend to advertise.
2

Pick the Venice models you'll resell

Examples: text-embedding-bge-m3, tts-kokoro, plus chat models supported by Venice. Check Venice’s docs for the exact apiUrl per model.
3

Stand up the proxy-router container

Follow Container P-Node. Start the container; it will create models-config.json defaults you can edit.
4

Edit models-config.json

{
  "$schema": "./internal/config/models-config-schema.json",
  "models": [
    {
      "modelId": "0x<your_chain_modelId_for_chat>",
      "modelName": "venice-chat",
      "apiType": "openai",
      "apiUrl": "https://api.venice.ai/api/v1/chat/completions",
      "apiKey": "<your_venice_api_key>",
      "concurrentSlots": 4,
      "capacityPolicy": "simple"
    },
    {
      "modelId": "0x<your_chain_modelId_for_embeddings>",
      "modelName": "text-embedding-bge-m3",
      "apiType": "openai",
      "apiUrl": "https://api.venice.ai/api/v1/embeddings",
      "apiKey": "<your_venice_api_key>"
    },
    {
      "modelId": "0x<your_chain_modelId_for_tts>",
      "modelName": "tts-kokoro",
      "apiType": "openai",
      "apiUrl": "https://api.venice.ai/api/v1/audio/speech",
      "apiKey": "<your_venice_api_key>"
    }
  ]
}
Restart the proxy-router after edits.
5

Register on chain

Same as a full provider, without the tee tag (you cannot attest Venice). Follow Register on chain.
6

Pick a bid price

Lower is more attractive but you must clear your Venice cost. See Registering a bid for the math.
7

Validate end-to-end

Open http://your-host:8082/swagger/index.html. From a separate consumer node, list models, open a session against your bid, and prompt. Watch your proxy-router logs and Venice usage dashboard simultaneously.

​
Operational tips

  • Track Venice usage — set Venice account limits/alerts so a session can’t blow your budget.
  • Throttle concurrentSlots — start conservatively. Going too high causes upstream 429s, which surface to consumers as a bad experience.
  • Failover: if you maintain multiple upstream accounts, run multiple proxy-routers and post separate bids; consumers route to whichever is cheapest and healthy.
  • Avoid the tee tag — you cannot prove anything about Venice’s runtime. Only use tee when you control the backend on a SecretVM-style TEE.