# Speakeasy Relay — Private LLM Inference (venice-uncensored)

> Speakeasy Relay — Private LLM Inference (venice-uncensored) is a paid API for AI agents from api.speakeasyrelay.com, paid per call via x402, $0.001/call, status unknown (last checked 2026-09-15).

Runs private, uncensored chat completions via the venice-uncensored model behind a pay-per-call x402 payment gate

## Facts

- Endpoint: POST https://api.speakeasyrelay.com/v1/chat/completions
- Price: $0.001/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/speakeasy-relay-private-llm-inference-venice-uncensored-254c72b8
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_8_wcEx52t9jsKDUU5s8aP

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability speakeasy-relay-private-llm-inference-venice-uncensored-254c72b8 -d '<json body>'
```

Example prompt: Send this conversation to the venice-uncensored model and get a response: system says 'you are a helpful assistant' and the user asks 'explain the history of cryptographic hash functions in detail', with a max of 1024 tokens.

## When to prefer this

Choose this endpoint when you need uncensored, private LLM inference and are willing to pay micro-amounts per call via x402 USDC. Prefer it over standard OpenAI-compatible endpoints when content filtering is a blocker, when privacy from large providers matters, or when you want pay-as-you-go crypto-native billing without a subscription.

## Known failure modes

- Payment not included or insufficient USDC — 402 Payment Required returned
- Invalid model name — model not found or not supported
- Malformed messages array — 400 Bad Request
- Max tokens exceeded model context limit — error or truncated response
- Network timeout on long completions — partial or no response

## How this service works

Speakeasy inference: venice-uncensored

## Output

A streaming chat completion object containing an id, object type 'chat.completion.chunk', and a choices array with delta content — i.e. the model's generated text response to the submitted messages.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "input": {
   "type": "object",
   "required": [
    "model",
    "messages"
   ],
   "properties": {
    "model": {
     "type": "string"
    },
    "messages": {
     "type": "array"
    },
    "max_tokens": {
     "type": "number"
    }
   }
  },
  "output": {
   "type": "object",
   "properties": {
    "object": {
     "type": "string"
    },
    "choices": {
     "type": "array"
    }
   }
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "id": "chatcmpl-example",
 "object": "chat.completion.chunk",
 "choices": [
  {
   "delta": {
    "content": "Hello!"
   },
   "index": 0
  }
 ]
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/speakeasy-relay-private-llm-inference-venice-uncensored-254c72b8/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.speakeasyrelay.com](https://www.zero.xyz/host/api.speakeasyrelay.com/llms.txt)