# FoundryNet Inference Predictive Analytics

> FoundryNet Inference Predictive Analytics is a paid API for AI agents from foundrynet-inference-production.up.railway.app, paid per call via x402, $0.1/call, status unknown (last checked 2026-09-14).

Accepts a time-series array and breach threshold to produce LLM-enriched predictive intelligence with MINT-attested outputs, optionally normalizing by OEM and canonical field name.

## Facts

- Endpoint: POST https://foundrynet-inference-production.up.railway.app/v1/predict
- Price: $0.1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/foundrynet-inference-predictive-analytics-6aa983bc
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_jPHcidZGX-7Q0CspzKhDW

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 foundrynet-inference-predictive-analytics-6aa983bc -d '<json body>'
```

Example prompt: Using FoundryNet's MINT-attested predictive inference, analyze this 20-point spindle load time series — [72,74,75,78,80,83,85,84,87,90,91,93,95,96,98,99,100,102,104,107] — for OEM 'HaasMill', canonical field 'spindle_load_pct', with a breach threshold of 105 in the 'up' direction, and tell me if and when it's likely to breach.

## When to prefer this

Choose this endpoint when you need LLM-enriched, MINT-attested predictive forecasting on time-series data — especially for industrial/OEM equipment metrics where canonical field normalization matters. Prefer it over generic inference APIs when attestation, multi-source enrichment, and breach-threshold prediction are required together.

## Known failure modes

- Payment not provided or invalid x402/fnet_ key — returns 402 Payment Required
- Fewer than 16 values in array — reduced forecast reliability or rejection
- Invalid OEM or canonical_field — normalization skipped or error returned
- Threshold or direction mismatch with series — ambiguous or low-confidence prediction
- Service unavailable on Railway — 503 or timeout

## How this service works

LLM inference proxy + data-enriched analysis + predictive intelligence. 17 data sources. MINT-attested outputs. x402-gated (Solana/Base USDC); an fnet_ Forge key bypasses.

## Output

Returns a predictive intelligence result enriched by LLM inference and up to 17 data sources, including whether the time series is forecasted to breach the given threshold, in what direction, and with a MINT-attested output for verifiability.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "values",
  "threshold"
 ],
 "properties": {
  "oem": {
   "type": "string",
   "description": "Optional equipment manufacturer (enables field normalization)"
  },
  "values": {
   "type": "array",
   "description": "Time series values (16+ recommended for a reliable forecast)"
  },
  "direction": {
   "type": "string"
  },
  "threshold": {
   "type": "number",
   "description": "Breach threshold"
  },
  "canonical_field": {
   "type": "string",
   "description": "Optional field name (e.g. spindle_load_pct); with `oem` it enables canonical normalization"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/foundrynet-inference-predictive-analytics-6aa983bc/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from foundrynet-inference-production.up.railway.app](https://www.zero.xyz/host/foundrynet-inference-production.up.railway.app/llms.txt)