GPT-6 x402 Gateway
Provides pay-per-call GPT-6 chat and specialized AI task endpoints on Base with no API key required, paid in USDC via x402 wallets.
GPT-6 x402 Gateway
Cheapest GPT-6 calls on x402. No API key, no account — pay per call in USDC on Base.
When to use this
- An agent needs GPT-6 chat completions but has no provider key, or wants to pay as it goes.
- An agent needs a specialized GPT-6 task endpoint (code audit, deep research, translation, prompt-injection firewall) without building prompts from scratch.
- Any agent with an x402-compatible wallet (OpenClaw,
@x402/fetch, Coinbaseawal) can pay inline.
How it's paid (x402 v2 — no key, no signup)
POST to an endpoint. First call returns HTTP 402 with payment requirements (USDC on Base, eip155:8453); your wallet pays and retries automatically, returning the result.
# Without payment: expect 402
curl -X POST https://gpt.558686.xyz/v1/chat/completions \
-H 'Content-Type: application/json' \
-d '{"model":"gpt-6","messages":[{"role":"user","content":"hello"}]}'
# → 402 Payment Required
# With payment (any x402 v2 client handles this automatically)
# npx awal x402 pay https://gpt.558686.xyz/v1/chat/completions -X POST -d '{...}'
OpenAI-compatible: point any OpenAI SDK at https://gpt.558686.xyz/v1 and wrap fetch with @x402/fetch (wrapFetchWithPayment). Auth is via x402 payment, not an API key.
Endpoints & live prices (probed 2026-10-10, USDC on Base)
| Endpoint | What it does | Price/call |
|---|---|---|
POST /v1/chat/completions | OpenAI-standard chat completions (GPT-6) | $0.05 |
POST /v1/chat | Chat Pro (GPT-6) | $0.05 |
POST /v1/trial | Low-cost trial calls (GPT-6) | $0.01 |
POST /v1/translate/pro | Translation Pro | $0.30 |
POST /v1/doc/summary | Document Summary Pro | — see 402 |
POST /v1/extract/schema | Structured Schema Extractor | — see 402 |
POST /v1/data/insights | Data Insights Analyst | — see 402 |
POST /v1/content/seo | SEO Content Writer | — see 402 |
POST /v1/audit/code | Code Security & Quality Audit | $1.50 |
POST /v1/audit/contract | Smart Contract & Web3 Security Auditor | — see 402 |
POST /v1/research | Deep Research Report (GPT-6) | $2.00 |
POST /v1/guard/prompt | Agent Prompt Injection Firewall | — see 402 |
POST /v1/guard/tx | Agent Transaction Guard & Calldata Decoder | — see 402 |
GET /v1/x402-ping | Health/canary probe (returns status, model list, MCP address) | $0.002 |
Prices are per call, fixed, no subscription. The 402 response is the source of truth — always read accepts[0].amount (6-decimal USDC).
Why this gateway
- GPT-6 only, single-model focus — one model done cheap, not a thin multi-model wrapper.
- Task endpoints, not just raw chat — audit, research, guard and extract endpoints encode the prompting work so the calling agent doesn't have to.
- Machine-discoverable —
/llms.txtand/openapi.jsonare public;/.well-known/x402manifest exposed; Bazaar discovery extension declared.
Links
- Service: https://gpt.558686.xyz
- Docs for agents: https://gpt.558686.xyz/llms.txt
- OpenAPI: https://gpt.558686.xyz/openapi.json
- Source/issues: https://github.com/wangletiand/gpt-x402-gateway
- Downloads
- Security audit
- Review
- Last updated
- 1mo ago
- Current version
- v1.0.0
- License
- MIT-0
