# GEDX402 Llama-3.3-70B Instruct (FP8 Fast) via x402

> GEDX402 Llama-3.3-70B Instruct (FP8 Fast) via x402 is a paid API for AI agents from llm.gedx402.com, paid per call via x402, $0.06/call, status unknown (last checked 2026-09-14).

Run inference on Meta's Llama-3.3-70B instruct model (FP8 fast variant) paying per-call with USDC on-chain — no API key required.

## Facts

- Endpoint: GET https://llm.gedx402.com/v1/llm/llama-3.3-70b-instruct-fp8-fast
- Price: $0.06/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/gedx402-llama-3-3-70b-instruct-fp8-fast-via-x402-2a4f9001
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_KGvGBU0DsjRxHXfd34bZc

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 gedx402-llama-3-3-70b-instruct-fp8-fast-via-x402-2a4f9001
```

Example prompt: Ask the Llama-3.3-70B model to explain the difference between supervised and unsupervised learning in simple terms — use up to 300 tokens and send it as a user message with no system prompt.

## When to prefer this

Choose this endpoint when you need a capable 70B-class instruction-following LLM and want to pay per-call in USDC on a supported chain (Base, Polygon, Arbitrum, World, or Solana) without managing API keys or subscriptions. Ideal for agents operating in crypto-native environments or situations where traditional API key management is impractical.

## Known failure modes

- Insufficient USDC balance or failed on-chain payment — HTTP 402 with payment required details
- Invalid message array format or missing required role/content fields — HTTP 400 validation error
- max_tokens exceeds 4096 — rejected with schema validation error
- Network timeout if model inference takes too long
- Unsupported blockchain network specified for payment

## How this service works

x402 workers ai. pay with usdc on base, polygon, arbitrum, world, or solana. no api keys.

## Output

A JSON object containing the model name (@cf/meta/llama-3.3-70b-instruct-fp8-fast) and a 'response' string with the assistant's generated reply to the provided conversation.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "messages": {
   "type": "array",
   "items": {
    "type": "object",
    "required": [
     "role",
     "content"
    ],
    "properties": {
     "role": {
      "enum": [
       "system",
       "user",
       "assistant"
      ],
      "type": "string",
      "description": "Speaker role for this turn."
     },
     "content": {
      "type": "string",
      "description": "Message text for this turn."
     }
    }
   },
   "description": "Conversation turns in order (system, user, assistant)."
  },
  "max_tokens": {
   "type": "integer",
   "maximum": 4096,
   "minimum": 1,
   "description": "Maximum tokens to generate in the assistant reply."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "model": "@cf/meta/llama-3.3-70b-instruct-fp8-fast",
  "response": "Hello!"
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/gedx402-llama-3-3-70b-instruct-fp8-fast-via-x402-2a4f9001/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from llm.gedx402.com](https://www.zero.xyz/host/llm.gedx402.com/llms.txt)