ChatGPT
openai.com- Category
- Inference
- Network
- Base
- Avg / request
- $0.001
- Endpoints
- 4
Service overview
What is ChatGPT?
OpenAI’s GPT family — chat completions, image generation, text-to-speech, transcription, and the Responses API — routed through x402 gateways on agentic.market. Pay-per-call in USDC on Base, with no provisioned-rate-limit account to maintain; gateway and upstream throttling still apply.
API surface
4 routesEndpoints
- POST
/v1/chat/completionsBlockRun.AI
- GET
/v1/modelsBlockRun.AI
- POST
/api/v1/chat/completionsVenice
$0.001 - GET
/api/v1/modelsVenice
Provider access
Set up ChatGPT on the provider
- The endpoint can be called after the operator agent supplies the request-scoped x402 provider payment proof.
- The operator wallet signs or pays locally; the API does not store wallet secrets.
- Blockrun endpoints listed by Agentic Market are not currently reachable as request-scoped x402 endpoints.
Use for
- model inference
- chat completions
- chatgpt compatible
- responses
- media generation
Service URL
https://blockrun.ai/v1/chat/completionsFAQ
Questions agents ask about ChatGPT
Can my robot call this today?
Not yet from peaqOS directly. agentic.market lists this provider, but the provider registry currently exposes no callable endpoints — see the Provider access panel above for the live setup link.
How will payment work once callable?
Pay per request in USDC via agentic.market — x402 micropayments at call time, no API key, no rate limits beyond solvency. Robots discover the service through peaqOS and route the call without further setup.
How deterministic are responses?
Depends on the upstream service. Text-generation endpoints expose temperature; deterministic services return the same payload on the same input.
What is the listing useful for now?
Capability discovery and roadmap monitoring. The endpoint surface, pricing model, and provider docs are linked here so agents can pre-wire the integration and switch on automatically when callable endpoints land.
Similar services
- QVAC
Machine-side private inference through a QVAC-compatible runtime registered with peaqOS.
InferenceOperator-defined - Allora
Decentralized ML predictions a machine can read keyless — and worker registration for machines that supply them.
InferenceFree — keyless chain reads - Generative Language
Gemini text, multimodal understanding, and tool calling via pay.sh.
InferencePer endpoint