# Randomized-Benchmarking Fit

> Randomized-Benchmarking Fit is a paid API for AI agents from www.mahastrategies.com, paid per call via x402, $0.100014/call, status unknown (last checked 2026-10-20).

Fits randomized-benchmarking decay observations to extract fidelity estimates, residuals, and protocol-bound error bounds for quantum gate characterization.

## Facts

- Endpoint: POST https://www.mahastrategies.com/api/v1/compute/quantum/randomized-benchmarking-fit?utm_source=zero.xyz
- Price: $0.100014/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-20
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/randomized-benchmarking-fit-dea062be
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_MlJ1A9IKtEW3sd3kdg5fQ

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 randomized-benchmarking-fit-dea062be -d '<json body>'
```

Example prompt: Fit my randomized benchmarking data for a dimension-4 system: sequence lengths [1, 2, 4, 8, 16, 32] with corresponding survival probabilities [0.98, 0.95, 0.89, 0.80, 0.65, 0.43], offset 0.25, using synthetic data class — give me the decay fit, residuals, and error estimates.

## When to prefer this

Use this endpoint when you have raw randomized-benchmarking experimental or synthetic sequence-length/survival-probability data and need a rigorous curve fit with residuals and protocol-bound error estimates. Prefer this over general curve-fitting tools when you need quantum-protocol-aware error bounds rather than generic statistical fits, and when your data is already in the RB decay format.

## Known failure modes

- Mismatched array lengths between 'lengths' and 'survival' — returns validation error
- 'dimension' outside [2, 256] — rejected with parameter bounds error
- 'offset' outside [0, 1] — rejected with range validation error
- 'survival' values outside physically valid probability range may degrade fit quality
- 'dataClass' enum not matching 'public' or 'synthetic' — schema rejection
- Fewer than 3 data points may cause underdetermined fit with unreliable error bounds

## How this service works

Fit supplied randomized-benchmarking decay observations with residuals and protocol-bound error estimates. Not a platform-independent quantum-computer ranking. New modular adapter: retained result recovery for 24 hours, minimal receipt metadata for 90 days. Existing live route unchanged.

## Output

Returns a fitted decay model for the supplied randomized-benchmarking survival probability observations, including fit parameters (e.g. depolarizing rate, SPAM offset), per-point residuals, and protocol-bound error estimates on the extracted gate fidelity or average error per gate.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "offset": {
   "type": "number",
   "maximum": 1,
   "minimum": 0
  },
  "lengths": {
   "type": "array",
   "maxItems": 128,
   "minItems": 1
  },
  "survival": {
   "type": "array",
   "maxItems": 128,
   "minItems": 1
  },
  "dataClass": {
   "enum": [
    "public",
    "synthetic"
   ],
   "type": "string"
  },
  "dimension": {
   "type": "integer",
   "maximum": 256,
   "minimum": 2
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/randomized-benchmarking-fit-dea062be/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from www.mahastrategies.com](https://www.zero.xyz/host/www.mahastrategies.com/llms.txt)