Cascade Inference
OpenAI-compatible chat completions gateway with 7 models: Kimi K2.5, MiniMax M2.5, Qwen 2.5 7B, Claude Sonnet 4.5/4.6, Claude Opus 4.5/4.6. Dynamic input-aware pricing from $0.001/req for cheap models to $0.40/req for Opus with large prompts. Supports streaming, max_tokens control. Dual-network x402 (Solana + Base USDC).
Start with a quote-only recipe. Payment signing stays in your project.
const response = await fetch("https://inference.surf.cascade.fyi/v1/chat/completions", {
method: "GET",
});
console.log(response.status);
console.log(Object.fromEntries(response.headers));
console.log((await response.text()).slice(0, 1000));
// A valid 402 is a quote, not a completed paid call.This sends no payment. Inspect the 402 response before adding a wallet-enabled client.
- Exact route checked
skipped origin unreachable · origin-check:UNSAFE_URL_ENOTFOUND: https://inference.surf.cascade.fyi/
- Origin unavailable
UNSAFE_URL_ENOTFOUND: https://inference.surf.cascade.fyi/
- Observed in 402-index
Discovered through satring.
- Registry record checked
Registry health: down.
EVIDENCE
Each record has an exact-route quote outcome. Unresolved templates and potentially mutating methods are labeled instead of being invoked without provider-specific test input. A quote is still separate from a settled paid call.