Benchmark Almanac

HEART

GPT-5 1,612 leads since January 2026
Not in model cards

Humans and models respond to people seeking emotional support. Judges compare responses and explain their preferences.

The paper compares human and model judgments.

Made by
Hippocratic AI
Released
January 2026
Kind
Wellbeing and sycophancy
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Measures
Preference
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Strongest model tested
Gemini 3 Pro #56 on the capability index
Last model added
November 2025
Scores refreshed
November 20th, 2026
15
Entries in its paper, 13 models we track

GPT-5 leads with 1,612 among models released by November 2025 GPT-5 leads with 1,612 among models released by November 2025

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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. This view shows only models we track, each at its best.

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

Bars start at 1,350.

  1. 1 GPT-5 1,612
  2. 2 o3 1,588
  3. 3 Claude Haiku 4.5 1,562
  4. 4 Kimi K2 Sep 2025 1,553
  5. 5 Claude Sonnet 4.5 1,541
  6. 6 GPT-4.1 mini 1,539
  7. 7 Claude 3.7 Sonnet 1,532
  8. 8 Claude Opus 4 1,504
  9. 9 GPT-4o (snapshot unspecified) 1,496
  10. 10 Gemini 2.5 Pro (snapshot unspecified) 1,478
  11. 11 Gemini 3 Pro 1,471
  12. 12 Llama 4 Maverick 1,441
  13. 13 Gemini 2.5 Flash Sep 2025 1,428
  14. 14 Llama 3.1-405B 1,415
  15. 15 Llama 3.1-70B 1,385

Source: The HEART paper

#EntryScoreSettingDateSource
1 GPT-5 1,612 Aug 2025 Official leaderboard
2 o3 1,588 Apr 2025 Official leaderboard
3 Claude Haiku 4.5 1,562 Oct 2025 Official leaderboard
4 Kimi K2 Sep 2025 1,553 Sep 2025 Official leaderboard
5 Claude Sonnet 4.5 1,541 Sep 2025 Official leaderboard
6 GPT-4.1 mini 1,539 Apr 2025 Official leaderboard
7 Claude 3.7 Sonnet 1,532 Feb 2025 Official leaderboard
8 Claude Opus 4 1,504 May 2025 Official leaderboard
9 GPT-4o (snapshot unspecified) 1,496 Official leaderboard
10 Gemini 2.5 Pro (snapshot unspecified) 1,478 Official leaderboard
11 Gemini 3 Pro 1,471 Nov 2025 Official leaderboard
12 Llama 4 Maverick 1,441 Apr 2025 Official leaderboard
13 Gemini 2.5 Flash Sep 2025 1,428 Sep 2025 Official leaderboard
14 Llama 3.1-405B 1,415 Jul 2024 Official leaderboard
15 Llama 3.1-70B 1,385 Jul 2024 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 1,612 Aug 2025 Official leaderboard
2 o3 1,588 Apr 2025 Official leaderboard
3 Claude Haiku 4.5 1,562 Oct 2025 Official leaderboard
4 Kimi K2 Sep 2025 1,553 Sep 2025 Official leaderboard
5 Claude Sonnet 4.5 1,541 Sep 2025 Official leaderboard
6 GPT-4.1 mini 1,539 Apr 2025 Official leaderboard
7 Claude 3.7 Sonnet 1,532 Feb 2025 Official leaderboard
8 Claude Opus 4 1,504 May 2025 Official leaderboard
9 Gemini 3 Pro 1,471 Nov 2025 Official leaderboard
10 Llama 4 Maverick 1,441 Apr 2025 Official leaderboard
11 Gemini 2.5 Flash Sep 2025 1,428 Sep 2025 Official leaderboard
12 Llama 3.1-405B 1,415 Jul 2024 Official leaderboard
13 Llama 3.1-70B 1,385 Jul 2024 Official leaderboard

Number one over time

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

OpenAI
Benchmark paper Not tested since Jan 2026
GPT-5 1,612

Source: The HEART paper

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
In the benchmark’s paper
1,400 1,500 1,600 2025 2026 1,400 1,500 1,600 2025 2026

Source: Official HEART 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

7 of 9 checks pass, 2 warnings. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (15 entries).
  • Pass Models scored 13 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 12 months ago, in November 2025, for Gemini 3 Pro.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3 Pro (#56).
  • Note Headroom Scored as an Elo rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 13 rows for models we track and 2 other entries 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