← x402station.io

ChatGPT

platform.openai.comInferenceprotocol: x402first seen 183d ago

GPT, Responses, Images, and audio APIs grouped across x402-enabled gateway providers.

Endpoints
9
Uptime · 24h
44.4%
48 / 108 probes
Avg latency · 24h
291 ms
p90 487 ms
Total calls
0
0 unique payers

Healthy probes · last 24h

7 hour buckets · hover for per-bucket stats

Endpoints

Ordered: active first, healthy first, cheapest first. Up to 200 shown.

EndpointMethodPriceNetworkLast statusLatencyProbed
https://api.venice.ai/api/v1/modelsGET—eip155:8453200333 ms33m ago
https://blockrun.ai/v1/modelsGET—eip155:8453200393 ms22m ago
https://blockrun.ai/v1/chat/completionsPOST—eip155:8453402179 ms33m ago
https://blockrun.ai/v1/images/generationsPOST—eip155:8453402180 ms30m ago
https://api.venice.ai/api/v1/images/generationsPOST0.001000 USDCeip155:8453400149 ms30m ago
https://api.venice.ai/api/v1/responsesPOST0.001000 USDCeip155:8453400262 ms22m ago
https://api.venice.ai/api/v1/audio/speechPOST0.001000 USDCeip155:8453400311 ms33m ago
https://api.venice.ai/api/v1/audio/transcriptionsPOST0.001000 USDCeip155:8453400508 ms33m ago
https://api.venice.ai/api/v1/chat/completionsPOST0.001000 USDCeip155:8453400142 ms31m ago