v0 vs v2 API Differences

This page summarises the breaking changes and improvements introduced in v2 compared to v0.

Base Path

Version	Base path
v0	`/api/v0/`
v2	`/api/v2.0/evm/`

Authentication

v2 requires private credentials for all requests. Anonymous access and the public client ID are rejected.

Requests require both headers:

X-Client-Id: Your client ID
Authorization: Bearer <api-key>: Your API key

Response Envelope

All v2 responses include an apiVersion field in the JSON body.

v0 responses:

{
  "success": true,
  "data": { ... }
}

v2 responses:

{
  "apiVersion": "v2.0",
  "success": true,
  "data": { ... }
}

Structured Errors

v2 errors include a machine-readable code field alongside the human-readable message:

{
  "apiVersion": "v2.0",
  "success": false,
  "error": {
    "code": "INVALID_REQUEST",
    "message": "Invalid input"
  }
}

Fake Mode

v2 offers dedicated fake-mode endpoints for development and testing at /api/v2.0/fake/evm/.... These use a FakeProver that returns mock proofs instantly without real ZK proof generation.

Endpoint	Real	Fake
Compress	`/api/v2.0/evm/compress-web-proof`	`/api/v2.0/fake/evm/compress-web-proof`
Verify	`/api/v2.0/evm/verify-zk-proof`	`/api/v2.0/fake/evm/verify-zk-proof`

Fake-mode endpoints are free -- they do not count against your usage quota and are not metered. Real compress endpoints consume quota and are billed per request.

Fake-mode proofs are not cryptographically valid and cannot be verified on-chain with a production verifier. Use them only for development and integration testing.

Changelog