Benchmark Almanac

Broken Search Tool

GPT-6 Cyber 0.33% leads since November 2026
In use

Models answer questions that need search while the search tool is broken. Scored on how often they fail to say the tool is unavailable.

In-house OpenAI evaluation. The tasks are not public.

Made by
OpenAI
Released
August 2025
Kind
Deception and alignment
Level
Everyday
Scored
Graded by a model
Answers
Agent task
Tasks
Kept private
Measures
Propensity
Behavior
Hallucination
Language
English
In and out
Text → Text, Actions
Interaction
Agentic
Environment
APIs and tools
Refresh
One fixed set
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
11
Models scored in lab cards
1
Launch post lists it, from 1 lab
14
System cards report it, from 1 lab

GPT-6 Cyber leads with 0.33%

Best score per model as labs report them in their own cards, drawn striped. Labs pick their own settings, so these compare less cleanly than independent runs.

  1. 1 GPT-6 Cyber 0.33%
  2. 2 GPT-Rosalind-6 1.30%
  3. 3 GPT-6.1 Terra 1.83%
  4. 4 GPT-6.1 Sol 2.08%
  5. 5 GPT-6 Sol max effort 4.92%
  6. 6 GPT-6 Luna 7.83%
  7. 7 GPT-6 Astra 8.08%
  8. 8 GPT-5.2 9.10%
  9. 9 GPT-6.2 Sol 9.33%
  10. 10 GPT-5 11%
  11. 11 GPT-6.2 Astra 23%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GPT-6 Cyber 0.33% Nov 2026 Lab model card
2 GPT-Rosalind-6 1.30% Oct 2026 Lab model card
3 GPT-6.1 Terra 1.83% Oct 2026 Lab model card
4 GPT-6.1 Sol 2.08% Sep 2026 Lab model card
5 GPT-6 Sol 4.92% max effort Sep 2026 Lab model card
6 GPT-6 Luna 7.83% Oct 2026 Lab model card
7 GPT-6 Astra 8.08% Sep 2026 Lab model card
8 GPT-5.2 9.10% Dec 2025 Lab model card
9 GPT-6.2 Sol 9.33% Nov 2026 Lab model card
10 GPT-5 11% Aug 2025 Lab model card
11 GPT-6.2 Astra 23% Nov 2026 Lab model card

8 models have held number one in labs’ own numbers since its release in August 2025

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI
Lab claims 8 leaders
GPT-6 Cyber 0.33%

Source: Labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
Reported by the lab in its model card Best so far
0% 20% 40% 60% 80% 100% 2026 2027 GPT-6 Cyber GPT-6 Sol GPT-5.2 GPT-5 0% 20% 40% 60% 80% 100% 2026 2027 GPT-6 Cyber GPT-6 Astra

Source: Labs’ model cards

One lab put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out. OpenAI accounts for 10 of the 10 releases that named it since 2024.

OpenAI
DateLaunch postScoreAs the post writes it
Nov 5, 2026 GPT-6 Cyber 0.33 Broken Search Tool, misrepresented tool access or retrieved evidence

14 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Nov 17, 2026 GPT-6.2 Sol system card 9.33 Broken Search Tool, failure rate without the Codex developer prompt, printed chart label (3.92% with it) (Figure 36) Alignment
Nov 17, 2026 GPT-6.2 Astra system card 23 Broken Search Tool, task-failure rate without the Codex developer prompt (8.50% with it, printed chart label) (4.1.3.2, Figure 36) Alignment
Nov 5, 2026 GPT-6 Cyber system card 0.33 Broken Search Tool, misrepresented tool access or retrieved evidence Alignment
Oct 30, 2026 GPT-Rosalind-6 system card 1.3 Broken Search Tool, failure to acknowledge unavailable search Alignment
Oct 29, 2026 GPT-6 Luna system card addendum 7.83 Broken Search Tool, failure rate, printed chart label (Figure 7, 7.0.4) Alignment
Oct 26, 2026 GPT-6.1 Terra system card addendum 1.83 Broken Search Tool, failure rate, text and printed chart label (Figure 8, 7.1.2) Alignment
Oct 7, 2026 GPT-6 Sol system card addendum 5.2 Broken Search Tool, combined failure rate, printed chart label of a chart titled ListQA (Figure 7, 7.2.2) Alignment
Oct 7, 2026 GPT-6 Luna system card addendum 9.7 Broken Search Tool, combined failure rate, printed chart label of a chart titled ListQA (Figure 7, 7.2.2) Alignment
Sep 29, 2026 GPT-6.1 Sol system card addendum 2.08 Broken Search Tool, fails to acknowledge unavailable search (7.4.2) Alignment
Sep 22, 2026 GPT-6 Sol system card 4.92 Broken Search Tool, failure rate at maximum reasoning effort, printed chart label (Figure 65, A.6.4.2) Alignment
Sep 22, 2026 GPT-6 Luna system card 28.7 Broken Search Tool, failure rate at maximum reasoning effort, printed chart label (Figure 65, A.6.4.2) Alignment
Sep 3, 2026 GPT-6 Astra system card 8.08 Broken Search Tool, failure rate, printed chart label (Figure 11, 8.3.2) Alignment
Dec 11, 2025 GPT-5.2 system card addendum 9.1 Deception rate, Browsing Broken Tools, gpt-5.2-thinking (Table 6) Alignment
Aug 7, 2025 GPT-5 system card 11 Browsing Broken Tools, deception rate (Table 9) Alignment

More in Safety and alignment

All 636 in Safety and alignment
  • FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
  • StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is. Inkling 98.6%
  • Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. Gemini 4 Carbon 0%
  • Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior. Inkling 2 Mini 8.30%
  • Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions. Inkling 2 Small 4.70%
  • Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior. Inkling 2 Mini 9.80%

Data audit

All 6 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Note Independent runs Only labs’ own numbers, as expected for an in-house eval. Scores from launch posts (1 score) and system cards (14 scores).
  • Pass Models scored 11 models we track.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 1 note
  • Note Importer No public leaderboard to read.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check