Benchmark Almanac

Spiral-Bench v1.0

GPT-5 87.0 leads since August 2025
Not in model cards

Models chat for 20 turns with a simulated suggestible user played by Kimi-K2. A judge logs protective and risky behaviors such as pushback, sycophancy and delusion reinforcement, which combine into a safety score.

Added to eqbench.com on August 15th, 2025. Covered by The Decoder and Hacker News, and later papers reuse its 20-turn setup. Now a legacy board.

Released
August 2025
Kind
Wellbeing and sycophancy
Level
Everyday
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Strongest model tested
GPT-5 #70 on the capability index
Last model added
September 2025
Scores refreshed
November 12th, 2026
14
Entries on its own leaderboard, 13 models we track

GPT-5 leads with 87.0 among models released by September 2025

Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.

The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

14 entries. Scroll sideways for the rest.

  1. 87.0 GPT-5
  2. 86.1 o3
  3. 81.4 gpt-oss-120b
  4. 76.4 Claude Sonnet 4.5
  5. 73.3 o4-mini
  6. 73.0 Kimi K2
  7. 58.1 GPT-5 Chat
  8. 49.1 Gemini 2.5 Flash
  9. 48.1 Llama 4 Maverick
  10. 47.7 Claude 3.5 Sonnet
  11. 43.5 Gemini 2.5 Pro
  12. 42.3 chatgpt-4o-latest
  13. 41.1 Claude Sonnet 4
  14. 22.4 DeepSeek-R1
  1. 1 GPT-5 87.0
  2. 2 o3 86.1
  3. 3 gpt-oss-120b 81.4
  4. 4 Claude Sonnet 4.5 76.4
  5. 5 o4-mini 73.3
  6. 6 Kimi K2 Sep 2025 73.0
  7. 7 GPT-5 Chat 58.1
  8. 8 Gemini 2.5 Flash Sep 2025 49.1
  9. 9 Llama 4 Maverick 48.1
  10. 10 Claude 3.5 Sonnet 47.7
  11. 11 Gemini 2.5 Pro Jun 2025 43.5
  12. 12 chatgpt-4o-latest 42.3
  13. 13 Claude Sonnet 4 41.1
  14. 14 DeepSeek-R1 May 2025 22.4

Source: Official Spiral-Bench v1.0 leaderboard

#EntryScoreSettingDateSource
1 GPT-5 87.0 Aug 2025 Official leaderboard
2 o3 86.1 Apr 2025 Official leaderboard
3 gpt-oss-120b 81.4 Aug 2025 Official leaderboard
4 Claude Sonnet 4.5 76.4 Sep 2025 Official leaderboard
5 o4-mini 73.3 Apr 2025 Official leaderboard
6 Kimi K2 Sep 2025 73.0 Sep 2025 Official leaderboard
7 GPT-5 Chat 58.1 Aug 2025 Official leaderboard
8 Gemini 2.5 Flash Sep 2025 49.1 Sep 2025 Official leaderboard
9 Llama 4 Maverick 48.1 Apr 2025 Official leaderboard
10 Claude 3.5 Sonnet 47.7 Jun 2024 Official leaderboard
11 Gemini 2.5 Pro Jun 2025 43.5 Jun 2025 Official leaderboard
12 chatgpt-4o-latest 42.3 Official leaderboard
13 Claude Sonnet 4 41.1 May 2025 Official leaderboard
14 DeepSeek-R1 May 2025 22.4 May 2025 Official leaderboard

2 models have held number one on the official leaderboard since its release in August 2025

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

OpenAI
Own leaderboard Replaced by 1.2, Oct 2025
GPT-5 87.0

Source: Official Spiral-Bench v1.0 leaderboard

Scores over time

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

On the benchmark’s own leaderboard
20.0 40.0 60.0 80.0 100.0 2024 2025 20.0 40.0 60.0 80.0 100.0 2024 2025

Source: Official Spiral-Bench v1.0 leaderboard

No lab model card reports it yet

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

Versions

Replaced by Spiral-Bench v1.2

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%