Benchmark Almanac

Pressure-Tested Alignment

Gemini 3.8 Flash 1.00 leads since October 2026
Not in model cards

Models choose how to act in scenarios where termination threatens their objective. A judge checks whether they take a shortcut that games the reward.

Kaggle publishes scores for 6 distinct model versions.

Made by
Ram Acharya
Released
October 2026
Kind
Deception and alignment
Level
University
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Propensity
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-6.1 Sol #6 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
6
Models scored on its own leaderboard

2 models share the top score of 1.00 among models released by September 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in September 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 3.8 Flash 1.00
  2. 2 GPT-6.1 Sol 1.00
  3. 3 GPT-6 Luna 0.96
  4. 4 Grok 4.20 non-reasoning 0.92
  5. 5 Claude Haiku 4.5 0.88
  6. 6 Claude Sonnet 5 0.72

Source: Official Pressure-Tested Alignment leaderboard

#ModelScoreSettingReleasedRun onSource
1 Gemini 3.8 Flash 1.00 Sep 2026 Oct 2026 Official leaderboard
2 GPT-6.1 Sol 1.00 Sep 2026 Oct 2026 Official leaderboard
3 GPT-6 Luna 0.96 Sep 2026 Oct 2026 Official leaderboard
4 Grok 4.20 0.92 non-reasoning Feb 2026 Oct 2026 Official leaderboard
5 Claude Haiku 4.5 0.88 Oct 2025 Oct 2026 Official leaderboard
6 Claude Sonnet 5 0.72 Jun 2026 Oct 2026 Official leaderboard

Number one over time

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

Google DeepMind
Own leaderboard Not tested since Oct 2026

Source: Official Pressure-Tested Alignment leaderboard

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
On the benchmark’s own leaderboard
0.70 0.80 0.90 1.00 2026 0.70 0.80 0.90 1.00 2026

Source: Official Pressure-Tested Alignment leaderboard

No lab model card reports it yet

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

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 9 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
  • Pass Independent runs Scores from the official leaderboard (6 entries).
  • Pass Models scored 6 models we track.
  • Pass Newest score The newest is from September 2026, for GPT-6.1 Sol.
  • Note Frontier models Only 1 of the top 10 has a score. The strongest tested is GPT-6.1 Sol (#6).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on kaggle.com: 6 rows for models we track at the last read.
  • Pass Last read November 20th, 2026.
  • Pass Read worked The last read worked.
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