Index onlineMerchant portal
Api Endpoint

Trust Verdict Benchmark

Benchmark Trust Verdict. Is an AI benchmark a trustworthy capability signal right now, or is it saturated, contaminated, or near its ceiling so a high score is a floor and not a differentiator? Returns a trust band and a 0-100 trust score per benchmark, fusing the registry contamination and saturation flags with the live spread of the top model scores (frontier compression), plus a down-weight recommendation and an alternative benchmark to use instead. Optional ?benchmark= or ?category= filter. AFTA-signed. The "should I trust this SOTA claim" call.

Open interactive test console

Offering

Provider
tensorfeed.ai
Method
GET
Offering type
Api Endpoint
Transport type
http
Category
Security
Delivery
Not provided
Pricing model
Not provided
x402 version
2

Endpoint context

URL
https://tensorfeed.ai/api/premium/benchmark-trust-verdict
MIME type
application/json
First seen
2026-10-16T01:59:37.852Z
Last seen
Tags
None supplied

Website

Payment pricing

x402 payment terms are the authoritative price for calling this endpoint.

Payment option 1
0.02 USDC
Atomic price: 20000 · Network: eip155:8453 · Asset: USDC · Scheme: exact
Payment option 2
0.02 USDC
Atomic price: 20000 · Network: solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp · Asset: USDC · Scheme: exact
Raw x402 terms
[
  {
    "scheme": "exact",
    "network": "eip155:8453",
    "asset": "0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913",
    "amount": "20000",
    "payTo": "0x549c82e6bfc54bdae9a2073744cbc2af5d1fc6d1"
  },
  {
    "scheme": "exact",
    "network": "solana:5eykt4UsFv8P8NJdTREpY1vzqKqZKvdp",
    "asset": "EPjFWdd5AufqSSqeM2qN1xzybapC8G4wEGGkZwyTDt1v",
    "amount": "20000",
    "payTo": "B8uYDm3snMCAUwt6NWTV3u7akcmd1AWzCXKQ1dDKWcFJ"
  }
]

Observed health

Status
Healthy
Latest latency
465 ms
Average latency
878 ms
HTTP status
402
Last checked
Reachability
100% over 15 checks

Latency is measured to the first response headers. A reachable result does not verify response accuracy or merchant identity.

Input schema

{
  "type": "http",
  "method": "GET",
  "queryParams": {
    "category": "code"
  }
}

Output schema

{
  "type": "json",
  "example": {
    "ok": true,
    "count": 1,
    "filter": {
      "category": "code",
      "benchmark": null
    },
    "billing": {
      "credits_charged": 1,
      "credits_remaining": 49
    },
    "verdicts": [
      {
        "id": "humaneval",
        "name": "HumanEval",
        "status": "saturated",
        "signals": {
          "models_scored": 0,
          "top_score_spread": null,
          "ceiling_proximity": "at_ceiling",
          "frontier_compression": "unknown"
        },
        "category": "code",
        "trust_band": "contaminated",
        "score_range": "0-100% pass@1",
        "trust_score": 12,
        "frontier_score": "~97%",
        "recommendation": "Down-weight scores on this benchmark (high training-contamination risk, marked saturated, frontier score is near the ceiling). A high score is closer to a capability floor than a differentiator. Prefer LiveCodeBench for current code signal.",
        "leaderboard_url": "https://paperswithcode.com/sota/code-generation-on-humaneval",
        "contamination_risk": "high"
      }
    ],
    "capturedAt": "2026-05-28T08:00:00Z"
  }
}