Using the AccelFury API

2024-06-10

Using the AccelFury API

Our unified interface covers proof generation and AI acceleration. Send hardware jobs using the call below:

Copy
curl -H "Authorization: Bearer <token>" -d '{"bitstream":"sha256.bit","input":"0x1234"}' https://accelfury.com/api/v1/jobs

You can also use our JavaScript SDK:

Copy
import { createJob } from '@accelfury/sdk'

const job = await createJob({
bitstream: 'sha256.bit',
input: '0x1234'
})
console.log(job.id)

For more advanced options, continue with our ZK SDK guide. Our docs include benchmarks comparing against Cysic's GPU-based flow so you can make an informed decision.

API client

When calling API routes from the frontend, use the typed helper located in webapp/services/apiClient.ts:

Copy
import { apiFetch } from '@/services/apiClient'
const result = await apiFetch('/api/contact', {
method: 'POST',
data: { name: 'Jane', email: 'jane@example.com' }
})

The helper automatically prefixes relative URLs with NEXT_PUBLIC_BASE_URL.

gRPC service

See our gRPC API guide for instructions on generating TypeScript clients and an example call.

WebSocket helpers

Real-time features use the lightweight wrapper in services/socket.ts:

Copy
import { Socket } from '@/services/socket'
const ws = new Socket<{ message: string }>('wss://accelfury.com/ws')
ws.on('message', (m) => console.log('update', m))