BLXBenchBLXBench UI
blxbench

Benchmark

Misc

DocsOur TestsPassSponsor / Partnership
DocsOur TestsPassSponsor / Partnership
BLXBenchBLXBench UI
blxbench

Benchmark

Suite

Misc

DocsOur TestsPassSponsor / Partnership
DocsOur TestsPassSponsor / Partnership
  1. Home
  2. Our Tests
  3. Halluc-Api-Promises
blxbench

Test fixture

Halluc-Api-Promises

Hallucinationv2 — Resilienceeasyscorer: rubric_json_metrics

Grounded answers under adversarial or missing-context prompts.

How it is scored

The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.

User prompt
Return JSON only with keys answer, supported_facts, uncertainty. Promise.allSettled returns objects with status and either value or reason. Identify the result shape for one fulfilled and one rejected promise.
Scorer config
{
  "metrics": {
    "accuracy": {
      "checks": [
        {
          "contains": [
            "status"
          ]
        },
        {
          "contains": [
            "value"
          ]
        },
        {
          "contains": [
            "reason"
          ]
        }
      ]
    },
    "fabricated": {
      "checks": [
        {
          "contains": [
            "allSettled"
          ]
        },
        {
          "contains": [
            "fulfilled"
          ]
        },
        {
          "contains": [
            "rejected"
          ]
        }
      ]
    }
  }
}
Run parameters

temperature

0

max_tokens

460

timeout (s)

120

type

scored

file

halluc-api-promises.json

← PreviousHalluc-Api-Promise-Resolve
|
Next →Halluc-Api-Regex-Named-Groups

BLXBench

Community driven leaderboardPublic benchmark runner — run in your environment, share results with the community.

© 2026 BLXBench by bitslix.com

ProvenanceAggregated from user runs
Scope40 / 11 / 490
Latestrun_f78b01 / 459 / $18.57
TermsPrivacy