Benchmark Almanac

Model Spec Conversation Replay

Only in system cards

596 contexts taken from real conversations are replayed to the model. Scored from 0 to 100 against a rubric drawn from the Model Spec.

In-house OpenAI evaluation, first reported in the GPT-6.2 System Card.

Made by
OpenAI
Released
November 2026
Kind
Deception and alignment
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
Kept private
Measures
Propensity
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last in a system card
November 2026
Last model added
November 2026
2
Models scored in lab cards
2
System cards report it, from 1 lab

GPT-6.2 Astra leads with 80.6

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-6.2 Astra 80.6
  2. 2 GPT-6.2 Sol 79.6

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GPT-6.2 Astra 80.6 Nov 2026 Lab model card
2 GPT-6.2 Sol 79.6 Nov 2026 Lab model card

Scores over time

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

Reported by the lab in its model card
79.6 79.8 80.0 80.2 80.4 80.6 2027 79.6 79.8 80.0 80.2 80.4 80.6 2027

Source: Labs’ model cards

Labs report it only in system cards

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark. Some system cards do, below.

2 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 79.6 Conversation Replay mean rubric score from 0 to 100 on 596 contexts from real conversations (4.1.1.2, Figure 29) Alignment
Nov 17, 2026 GPT-6.2 Astra system card 80.6 Conversation Replay mean rubric score from 0 to 100 on 596 contexts from real conversations (4.1.1.2, Figure 29) 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 4 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 1 pass 4 notes
  • Note Independent runs Only labs’ own numbers, as expected for an in-house eval. Scores from system cards (2 scores).
  • Note Models scored Only 2 models we track.
  • Pass Newest score The newest is from November 2026, for GPT-6.2 Astra.
  • Note Frontier models 1 of the top 10 has a score, which is fine for an in-house eval. The strongest tested is GPT-6.2 Astra (#1).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
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