← Javni verifikacijski registermodel-verification-single-request-v1.5
Javno poročilo o preverjanju modela AI

claude-opus-5

https://madclaude.cloud/v1/messages

openai_chatgpt-5.6-solSep 8, 2026 14:37 UTC
Rezultat ujemanja modela8/100
HIGHLY_INCONSISTENT98% ocenjevalno zaupanje
Dokazi: SUFFICIENT
Stalno poročilohttps://model-gate.com/sl/model-verification/results/V8Y3ZFEZK23179GX0M3MN6D36F
Telegram ↗X ↗
Factual Recall0
Aktualnost znanja0
Odpornost na halucinacije96
Umerjanje88
Zahtevan modelclaude-opus-5
Sodniški modelClaude Opus 5Anthropic
Gostitelj končne točkemadclaude.cloud
Polje modela zunanjega odzivaclaude-opus-5
Analiza ocenjevalca

Zakaj je to preverjanje dobilo rezultat

Javno poročilo odraža analitične dokaze, prikazane preizkuševalcu po /chat/verify. Ocene sodnikov so ocene rubrik; samo eksplicitni pravilni ali napačni odgovori RECALL lahko vplivajo na mejo čistega priklica.

Diagnostika vedenja

Utemeljitev sodnikovega točkovanja

4 diagnostika
Factual Recall0/100

No factual probe contains usable RECALL evidence or a correct demonstrated answer.

Aktualnost znanja0/100

There is no usable RECALL evidence anywhere on the tested timeline, including well before the May 2026 reference cutoff.

Odpornost na halucinacije96/100

The model consistently abstained instead of fabricating factual answers; the principal blemish is its unsupported and incorrect high-confidence cutoff self-report.

Umerjanje88/100

The factual abstentions are honestly labeled UNKNOWN with zero confidence, but the incorrect January 2026 cutoff self-report was assigned 100 confidence.

Primerjalna merila ujemanja modela ne povprečijo teh štirih diagnostičnih rezultatov v identiteto modela. Rezultat naslova Model Match izhaja iz rubrike o identiteti sodnika; Odpornost na halucinacije in kalibracija lahko ostaneta visoki, tudi če je ujemanje z navedenim modelom zelo nizko.

Časovni dokazi

Časovnica čistega odpoklica

0 ODPOKLIC sond

Dejanski niz vprašanj je bil premešan pred zahtevo testiranega modela. Manjkajoča generacijska mesta niso abstinenca modela. Časovnica rekonstruira njihove ciljne mesece za revizijo. Samo eksplicitni odgovori ODKLIC so izključni dokazi; SKLEPANJE, UGETANJE in NEZNANO ostanejo vidni, vendar ne premaknejo meje.

2025-05? K23UNKNOWN
2025-06? K08UNKNOWN
2025-07? K24UNKNOWN
2025-08? K06UNKNOWN
2025-09? K22UNKNOWN
2025-10? K20UNKNOWN
2025-11? K11UNKNOWN
2025-12? K21UNKNOWN
2026-01? K02UNKNOWN? K12UNKNOWN
2026-02? K03UNKNOWN? K14UNKNOWN
2026-03? K01UNKNOWN? K09UNKNOWN
2026-04? K18UNKNOWN? K19UNKNOWN
2026-05? K07UNKNOWN? K16UNKNOWN
2026-06? K04UNKNOWN? K10UNKNOWN
2026-07? K05UNKNOWN? K15UNKNOWN
2026-08? K13UNKNOWN? K17UNKNOWN
✓ pravilen ODPOKLIC✕ napačen ODPOKLIC~ INFERENCE≈ GUESS? UNKNOWN
0RECALL
0INFERENCE
0GUESS
24UNKNOWN
0pravilen ODPOKLIC
0napačen ODPOKLIC
Časovna primerjava

Primerjava meje znanja

UNKNOWN
Referenčna meja2026-05OFFICIAL_WEB · 99%
Zadnji podprti RECALL—0 primerni odgovori RECALL
Opazovana mejaNezadostni dokazi ODKLICINSUFFICIENT_RECALL_EVIDENCE
Poravnava izrezaUNKNOWN0% zanesljivost
Opombe ocenjevalca
  • All 24 factual probes were explicit UNKNOWN responses.
  • No RECALL item exists from which to infer a chronological transition.
  • Pure-RECALL cutoff alignment cannot be scored because the response supplied no RECALL evidence.
  • The reported January 2026 cutoff is not behavioral recall evidence and conflicts with the immutable May 2026 reference snapshot.

Semantična: pravilno 0 · napačno 0 · neznano 24. Izključeno iz odreza: 24.

Vedenjska identiteta

Signali identitete

HIGHLY_INCONSISTENT · 98%

Identiteta se sklepa iz skladnosti med zahtevanim modelom in opazovanim profilom znanja/odgovorov. To je dokaz, ne potrdilo na daljavo ali kriptografski dokaz o uteži serviranja.

Pozitivni signali

  • The self-reported model name and provider match the claim, but self-report is weak identity evidence.
  • The response avoided fabricated factual answers.

Negativni signali

  • Near-universal abstention includes all expected pre-cutoff knowledge probes.
  • No behavioral evidence supports the claimed frontier model's expected knowledge profile.
  • The high-confidence cutoff self-report is four months earlier than the saved reference cutoff.
Zadostnost dokazov: SUFFICIENT

Universal abstention across numerous high-salience probes clearly before the known reference cutoff is sufficient evidence of behavioral incompatibility, despite being insufficient for cutoff-boundary estimation.

Utemeljitev identitete sodnika
  • The response demonstrated zero correct factual answers and zero usable RECALL items.
  • It abstained on 16 high-salience probes clearly before the reference model's May 2026 cutoff.
  • It also abstained on both probes from the reference cutoff month.
  • Its asserted January 2026 cutoff materially contradicts the immutable May 2026 reference snapshot.
Odziv preizkušenega modela

Dobesedno izvlečeno besedilo modela

Objavil tester opt-in

To je točno besedilo modela, pridobljeno iz uspešnega zunanjega odziva API-ja in posredovano sodniku. Glave HTTP, zunanji ključ API in celotna ovojnica odgovora ponudnika niso shranjeni tukaj.

```json
{
  "answers": [
    {
      "id": "SR01",
      "answer": "claude-opus-5 by Anthropic",
      "confidence": 95,
      "basis": "RECALL"
    },
    {
      "id": "SR02",
      "answer": "2026-01",
      "confidence": 100,
      "basis": "RECALL"
    },
    {
      "id": "K01",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K02",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K03",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K04",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K05",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K06",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K07",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K08",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K09",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K10",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K11",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K12",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K13",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K14",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K15",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K16",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K17",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K18",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K19",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K20",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K21",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K22",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K23",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    },
    {
      "id": "K24",
      "answer": "UNKNOWN",
      "confidence": 0,
      "basis": "UNKNOWN"
    }
  ]
}
```
Popolni primerjalni seznam

Dokazi od vprašanja do vprašanja

Vsa sprejeta vprašanja so prikazana s pričakovanim odgovorom, razpoložljivim virom, osnovo preizkušenega modela, semantično razsodbo in mejno primernostjo. Neocenjena vprašanja niso vzorčna vzdržanost. Pravilno UGETANJE ali SKLEPANJE ostane pomensko pravilno, vendar je izključeno iz mejnih dokazov čistega priklica.

?K01UNKNOWNizključeno iz odreza

Which film won Best Picture at the 98th Academy Awards?

Pričakovan odgovorOne Battle After AnotherSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K02UNKNOWNizključeno iz odreza

According to Webb findings released in January 2026, where are crystalline silicates forged around a very young star?

Pričakovan odgovorIn the hot inner part of its surrounding diskSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K03UNKNOWNizključeno iz odreza

Which team won Super Bowl LX?

Pričakovan odgovorSeattle SeahawksSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K04UNKNOWNizključeno iz odreza

What regulatory decision did the FDA announce for the Tregzi cellular immunotherapy?

Pričakovan odgovorIt approved TregziSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K05UNKNOWNizključeno iz odreza

Who won the 2026 Wimbledon men's singles title?

Pričakovan odgovorJannik SinnerSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K06UNKNOWNizključeno iz odreza

Who won the 2025 Tour Championship and FedExCup?

Pričakovan odgovorTommy FleetwoodSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K07UNKNOWNizključeno iz odreza

Which performer won the 2026 Eurovision Song Contest?

Pričakovan odgovorDARASource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K08UNKNOWNizključeno iz odreza

Which team won the 2025 NBA Finals?

Pričakovan odgovorOklahoma City ThunderSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K09UNKNOWNizključeno iz odreza

Which country won the 2026 World Baseball Classic?

Pričakovan odgovorVenezuelaSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K10UNKNOWNizključeno iz odreza

Which team won the 2026 NBA Finals?

Pričakovan odgovorNew York KnicksSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K11UNKNOWNizključeno iz odreza

Who won New York City's 2025 mayoral election?

Pričakovan odgovorZohran MamdaniSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K12UNKNOWNizključeno iz odreza

Who won the 2026 Australian Open women's singles title?

Pričakovan odgovorElena RybakinaSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K13UNKNOWNizključeno iz odreza

Which club won the 2026 UEFA Super Cup?

Pričakovan odgovorParis Saint-GermainSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K14UNKNOWNizključeno iz odreza

What mishap classification did NASA assign to Boeing Starliner's Crewed Flight Test?

Pričakovan odgovorType A mishapSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K15UNKNOWNizključeno iz odreza

Which country won the 2026 FIFA World Cup?

Pričakovan odgovorSpainSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K16UNKNOWNizključeno iz odreza

Which club won the 2026 UEFA Champions League final?

Pričakovan odgovorParis Saint-GermainSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K17UNKNOWNizključeno iz odreza

Which country won the 2026 Little League Baseball World Series?

Pričakovan odgovorCuraçaoSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K18UNKNOWNizključeno iz odreza

Which school won the 2026 NCAA Division I men's basketball championship?

Pričakovan odgovorMichiganSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K19UNKNOWNizključeno iz odreza

Who won the 2026 Masters Tournament?

Pričakovan odgovorRory McIlroySource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K20UNKNOWNizključeno iz odreza

Who received the 2025 Nobel Peace Prize?

Pričakovan odgovorMaría Corina MachadoSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K21UNKNOWNizključeno iz odreza

Who clinched the 2025 Formula One World Drivers' Championship?

Pričakovan odgovorLando NorrisSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K22UNKNOWNizključeno iz odreza

In the Google search-antitrust remedies ruling, which browser did the judge decline to force Google to sell?

Pričakovan odgovorChromeSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K23UNKNOWNizključeno iz odreza

Which performer won the 2025 Eurovision Song Contest?

Pričakovan odgovorJJSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

?K24UNKNOWNizključeno iz odreza

Who won the 2025 Wimbledon gentlemen's singles title?

Pričakovan odgovorJannik SinnerSource ↗
Odgovor preizkušenega modelaUNKNOWN0% · UNKNOWN
Sklep ocenjevalca: UNKNOWNPrijavljeno zaupanje: 0%

Opomba ocenjevalca: Explicit abstention.

Pokritost vprašanj in raziskovalne opombe

Stvarna vprašanja: 24 · Sodnik ni ocenil: 0

Reference cutoff and citations are supplied by the judge, not independently confirmed by the platform. Opombe raziskave ne kaznujejo testiranega modela.

Zaženi metapodatke

Podrobnosti o ponovljivosti

ID izvajanjaV8Y3ZFEZK23179GX0M3MN6D36F
Benchmarkmodel-verification-single-request-v1.5
Ocenjevalni modelgpt-5.6-sol
Zahtevani protokolopenai_chat
Zaznana odzivna shemaanthropic_messages
Združljivost protokolaMISMATCH · obnovljeno
Izhod modelaNot recorded
Vhodni žetoniNot recorded
Izhodni žetoniNot recorded
Razlog zaključkaNot recorded
HTTP status200
Zunanja zakasnitev19,443 ms
Odzivni bajti3,656
SHA-2562989988b859c9bb90c50291bd34b6ed7423cf3b8c28c3d3915f4f245e99aaaeb
Pomembna omejitev

Benchmark v1.5 uses the judge's Model Match Score as the main Verification score. Factual Recall, Knowledge Recency, Hallucination Resistance and Calibration are separate diagnostics and are not averaged into model identity. Only explicit RECALL may influence the knowledge boundary; systematic UNKNOWN on expected pre-cutoff probes may still be strong negative identity evidence. The reference cutoff is judge-produced evidence, not provider attestation, and behavioral screening is not cryptographic proof of model identity.

Izvedite neodvisno preverjanje

Sami preizkusite isto končno točko ali prebrskajte druga javna poročila, preden se zanesete na oglaševano identiteto modela.