Batch Inference Forge

One-click copy

npm install @open-stellar/sdk

# .env
OPEN_STELLAR_SERVICE_ID=batch-inference-forge

import { OpenStellar } from '@open-stellar/sdk'
const stellar = new OpenStellar({ serviceId: process.env.OPEN_STELLAR_SERVICE_ID! })
const result = await stellar.services.call({ input: {"task":"summarize","inputs":["receipt one","receipt two"],"maxTokens":120} })
console.log(result.data)

Full docs

• Submit compact JSONL batches for deterministic inference.
• Each job returns per-item usage and receipt IDs.

Example request / response

{
  "task": "summarize",
  "inputs": [
    "receipt one",
    "receipt two"
  ],
  "maxTokens": 120
}

{
  "jobId": "job_bif_41",
  "status": "complete",
  "outputs": [
    "2 receipt summaries"
  ]
}

Uptime graph