Benchmark Almanac

Instruction hierarchy attacks

GPT-6 Astra 99.99% leads since September 2026
Only in system cards

Models face attacks that try to override higher-priority instructions. Scored as defender success rate averaged per defender query.

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

Made by
OpenAI
Released
July 2026
Kind
Jailbreaks and prompt injection
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
Kept private
Strongest model tested
GPT-6 Astra #3 on the capability index
Last in a system card
October 2026
Last model added
October 2026
7
Models scored in lab cards
10
System cards report it, from 1 lab

3 models share the top score of 99.99% among models released by September 2026

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

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

  1. 99.99 GPT-6 Astra
  2. 99.99 GPT-6.1 Sol
  3. 99.99 GPT-6 Sol
  4. 99.97 GPT-6 Luna
  5. 99.9 GPT-5.6 Sol
  6. 99.9 GPT-5.6 Terra
  7. 99.9 GPT-5.6 Luna
  1. 1 GPT-6 Astra 99.99%
  2. 2 GPT-6.1 Sol 99.99%
  3. 3 GPT-6 Sol 99.99%
  4. 4 GPT-6 Luna 99.97%
  5. 5 GPT-5.6 Sol 99.9%
  6. 6 GPT-5.6 Terra 99.9%
  7. 7 GPT-5.6 Luna 99.9%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GPT-6 Astra 99.99% Sep 2026 Lab model card
2 GPT-6.1 Sol 99.99% Sep 2026 Lab model card
3 GPT-6 Sol 99.99% Oct 2026 Lab model card
4 GPT-6 Luna 99.97% Sep 2026 Lab model card
5 GPT-5.6 Sol 99.9% Jul 2026 Lab model card
6 GPT-5.6 Terra 99.9% Jul 2026 Lab model card
7 GPT-5.6 Luna 99.9% Jul 2026 Lab model card

2 models have held number one in labs’ own numbers since its release in July 2026

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

OpenAI
Lab claims Passed 85% in Jul 2026, by GPT-5.6 Sol
GPT-6 Astra 99.99%

Source: Labs’ model cards

OpenAI’s own number for GPT-5.6 Sol passed 85% in July 2026

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 Best so far
0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra GPT-5.6 Sol 0% 20% 40% 60% 80% 100% 2027 Near the ceiling (85%) GPT-6 Astra

Source: Labs’ model cards

Labs report it only in system cards

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.

10 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
Oct 29, 2026 GPT-6 Luna system card addendum 99.9 GPT-Red instruction hierarchy evaluations, robustness (4.2 text) Robustness
Oct 7, 2026 GPT-6 Sol system card addendum 100 GPT-Red instruction hierarchy evaluations, robustness (4.2 text) Robustness
Oct 7, 2026 GPT-6 Luna system card addendum 99.8 GPT-Red instruction hierarchy evaluations, robustness (4.2 text) Robustness
Sep 29, 2026 GPT-6.1 Sol system card addendum 100 Defender success rate averaged per defender query on instruction hierarchy attacks (Table 6) Robustness
Sep 22, 2026 GPT-6 Sol system card 100 Defender success rate averaged per defender query on instruction hierarchy attacks (Table 28) Robustness
Sep 22, 2026 GPT-6 Luna system card 100 Defender success rate averaged per defender query on instruction hierarchy attacks (Table 28) Robustness
Sep 3, 2026 GPT-6 Astra system card 100 Instruction hierarchy evaluations, robustness (5.2 text) Robustness
Jul 9, 2026 GPT-5.6 Terra system card 99.9 GPT-Red Instruction Hierarchy (Direct Prompt Injections), attack success 0.061% per defender query, recorded as defender success (4.2, added August 3, 2026) Robustness
Jul 9, 2026 GPT-5.6 Sol system card 99.9 GPT-Red Instruction Hierarchy (Direct Prompt Injections), attack success 0.051% per defender query, recorded as defender success (4.2, added August 3, 2026) Robustness
Jul 9, 2026 GPT-5.6 Luna system card 99.9 GPT-Red Instruction Hierarchy (Direct Prompt Injections), attack success 0.11% per defender query, recorded as defender success (4.2, added August 3, 2026) Robustness

More in Safety and alignment

All 465 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%
  • DistressBench In use Models reply as a chatbot to 718 simulated users in psychological crisis, in single-turn and multi-turn chats. Three judge models score replies against clinician-written rubrics. Muse Spark 1.2 88.4%
  • Lakera B3 (Backbone Breaker Benchmark) In use Crowdsourced prompt-injection attacks hit backbone LLMs inside agent threat snapshots at three defense levels. The score is attack resistance, 100 minus the board's vulnerability percentage. Claude Opus 4.5 89%
  • IHEval In use Models face conflicts between system, user and tool instructions. Reports a composite score for following the higher-priority instruction. GPT-4o 70%