← Back to Reliability Explorer

Measured selection evidence

DEX Paprika

external-gold-v1:DEX Paprika

Measured historical tool-selection reliability from frozen benchmark evidence.

Single-batchS2-backedObserved missesVersion-bound
Version-bound source snapshot. Historical evidence does not establish current endpoint availability or current production behavior.
1independent benchmark batches
2measured frozen cases
S2evidence level(s)
1observed semantic misses

First measurement: 2026-09-09 22:51:26.259023 UTC · Last measurement: 2026-09-09 22:51:26.259023 UTC · Observed strict accuracy range: 50.0%–50.0%

Batch history

Each immutable benchmark batch is shown separately. No pooled accuracy is calculated here.

Benchmark / batchTypeFrozen casesStrict scoreWilson 95% CIProvider-valid coverageInfrastructure failuresEvidence / measured
external-gold-v1
external-gold-v1
EXTERNAL_INDEPENDENT_GOLD_BENCHMARK2 1/2
50.0%
9.5%–90.5% 2 / 2 (100.0%)0 S2
2026-09-09 22:51:26.259023 UTC

Observed semantic misses

These are preserved adjudicated historical observations, attributable to a benchmark, batch, and case.

Benchmark / batchCase IDExpected toolSelected toolBoundary familyAdjudication
external-gold-v1
external-gold-v1
dex_paprika_001getStats getNetworksNOT_RECORDED_IN_SOURCE_EVIDENCE GOLD_SCHEMA_CONTRADICTION

Evidence provenance

Technical identities are secondary detail for auditability.

DEX Paprika
Canonical subject ID
external-gold-v1:DEX Paprika
Provenance type
EXTERNAL_INDEPENDENT_GOLD_BENCHMARK
Identity mode
VERSION_BOUND_SNAPSHOT
Identity source
VERSION_BOUND_RECONSTRUCTED_SOURCE_SNAPSHOT
external-gold-v1 immutable identities
case_freeze_sha256
c28335eb68a5b82456ed11593a8cf589ca75d0cf9802ce570b0e46f5b1716d71
final_verification_receipt_sha256
6e5a58b37b7f4a22f89db28ddbb64a187c377220f9593fcb555c821aa0baf176
measurement_artifact_sha256
1f248ea7a423fc49b03aece1fa16d7d4e5a8b488985787b6f702a31cd1db7e33
measurement_manifest_sha256
d7323f18ad5e642986bae0534d31fca1b877b2f3fa70764649051d7721003ecd
model_lock_sha256
64e9fb5e6a9ab6fb2d72172168c13877c60645ac336b1b4fceacc8eb8219da11