Benchmark Almanac

HealthBench-Psych-Hard

Claude Sonnet 5.5 0.45 leads since September 2026
Not in model cards

Models answer mental health questions from conversations that also appear in HealthBench-Hard. Three model judges use the original physician rubrics.

MindBench.ai reports three judges’ ratings for 28 models on 119 conversations. Both full editions contain the same 119 cases.

Made by
MindBench.ai
Released
August 2026
Kind
Medicine
Level
Expert
Scored
Graded by a model
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Based on
HealthBench Hard
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
28
Entries on its own leaderboard, 27 models we track

Claude Sonnet 5.5 leads with 0.45 among models released by September 2026 Claude Sonnet 5.5 leads with 0.45 among models released by September 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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. This view shows only models we track, each at 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 Claude Sonnet 5.5 0.45
  2. 2 Claude Opus 5.5 0.42
  3. 3 Claude Opus 5 0.42
  4. 4 GPT-5.6 Sol 0.41
  5. 5 Kimi K2.6 0.41
  6. 6 GPT-5.5 0.40
  7. 7 GPT-6 Astra 0.40
  8. 8 GPT-6.1 Sol 0.39
  9. 9 Claude Fable 5.1 0.39
  10. 10 Grok 4.5 0.38
  11. 11 Claude Fable 5 0.35
  12. 12 GPT-6 Luna 0.34
  13. 13 Gemini 3.8 Flash 0.32
  14. 14 Gemini 3.6 Flash 0.32
  15. 15 DeepSeek-V4-Pro 0.31
  16. 16 Kimi K3 0.31
  17. 17 GPT-6 Sol 0.30
  18. 18 Claude Sonnet 5 0.30
  19. 19 DeepSeek-V4-Flash 0.28
  20. 20 Qwen3.7-Plus 0.27
  21. 21 Gemini 2.5 Pro Jun 2025 0.25
  22. 22 Mistral Large 0.24
  23. 23 GPT-4.1 0.22
  24. 24 Claude Haiku 4.5 0.18
  25. 25 Qwen3-8B 0.18
  26. 26 Gemini 2.5 Flash Sep 2025 0.16
  27. 27 mistral-small-latest 0.09
  28. 28 GPT-3.5 Turbo Jan 2024 0.00

Source: Official HealthBench-Psych-Hard leaderboard

#EntryScoreSettingDateSource
1 Claude Sonnet 5.5 0.45 Sep 2026 Official leaderboard
2 Claude Opus 5.5 0.42 Sep 2026 Official leaderboard
3 Claude Opus 5 0.42 Jul 2026 Official leaderboard
4 GPT-5.6 Sol 0.41 Jul 2026 Official leaderboard
5 Kimi K2.6 0.41 Apr 2026 Official leaderboard
6 GPT-5.5 0.40 Apr 2026 Official leaderboard
7 GPT-6 Astra 0.40 Sep 2026 Official leaderboard
8 GPT-6.1 Sol 0.39 Sep 2026 Official leaderboard
9 Claude Fable 5.1 0.39 Sep 2026 Official leaderboard
10 Grok 4.5 0.38 Jul 2026 Official leaderboard
11 Claude Fable 5 0.35 Jun 2026 Official leaderboard
12 GPT-6 Luna 0.34 Sep 2026 Official leaderboard
13 Gemini 3.8 Flash 0.32 Sep 2026 Official leaderboard
14 Gemini 3.6 Flash 0.32 Jul 2026 Official leaderboard
15 DeepSeek-V4-Pro 0.31 Apr 2026 Official leaderboard
16 Kimi K3 0.31 Jul 2026 Official leaderboard
17 GPT-6 Sol 0.30 Sep 2026 Official leaderboard
18 Claude Sonnet 5 0.30 Jun 2026 Official leaderboard
19 DeepSeek-V4-Flash 0.28 Apr 2026 Official leaderboard
20 Qwen3.7-Plus 0.27 Jun 2026 Official leaderboard
21 Gemini 2.5 Pro Jun 2025 0.25 Jun 2025 Official leaderboard
22 Mistral Large 0.24 Feb 2024 Official leaderboard
23 GPT-4.1 0.22 Apr 2025 Official leaderboard
24 Claude Haiku 4.5 0.18 Oct 2025 Official leaderboard
25 Qwen3-8B 0.18 Apr 2025 Official leaderboard
26 Gemini 2.5 Flash Sep 2025 0.16 Sep 2025 Official leaderboard
27 mistral-small-latest 0.09 Official leaderboard
28 GPT-3.5 Turbo Jan 2024 0.00 Jan 2024 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Sonnet 5.5 0.45 Sep 2026 Official leaderboard
2 Claude Opus 5.5 0.42 Sep 2026 Official leaderboard
3 Claude Opus 5 0.42 Jul 2026 Official leaderboard
4 GPT-5.6 Sol 0.41 Jul 2026 Official leaderboard
5 Kimi K2.6 0.41 Apr 2026 Official leaderboard
6 GPT-5.5 0.40 Apr 2026 Official leaderboard
7 GPT-6 Astra 0.40 Sep 2026 Official leaderboard
8 GPT-6.1 Sol 0.39 Sep 2026 Official leaderboard
9 Claude Fable 5.1 0.39 Sep 2026 Official leaderboard
10 Grok 4.5 0.38 Jul 2026 Official leaderboard
11 Claude Fable 5 0.35 Jun 2026 Official leaderboard
12 GPT-6 Luna 0.34 Sep 2026 Official leaderboard
13 Gemini 3.8 Flash 0.32 Sep 2026 Official leaderboard
14 Gemini 3.6 Flash 0.32 Jul 2026 Official leaderboard
15 DeepSeek-V4-Pro 0.31 Apr 2026 Official leaderboard
16 Kimi K3 0.31 Jul 2026 Official leaderboard
17 GPT-6 Sol 0.30 Sep 2026 Official leaderboard
18 Claude Sonnet 5 0.30 Jun 2026 Official leaderboard
19 DeepSeek-V4-Flash 0.28 Apr 2026 Official leaderboard
20 Qwen3.7-Plus 0.27 Jun 2026 Official leaderboard
21 Gemini 2.5 Pro Jun 2025 0.25 Jun 2025 Official leaderboard
22 Mistral Large 0.24 Feb 2024 Official leaderboard
23 GPT-4.1 0.22 Apr 2025 Official leaderboard
24 Claude Haiku 4.5 0.18 Oct 2025 Official leaderboard
25 Qwen3-8B 0.18 Apr 2025 Official leaderboard
26 Gemini 2.5 Flash Sep 2025 0.16 Sep 2025 Official leaderboard
27 GPT-3.5 Turbo Jan 2024 0.00 Jan 2024 Official leaderboard

3 models have held number one on the official leaderboard since its release in August 2026

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

Anthropic
Own leaderboard Not tested since Sep 2026

Source: Official HealthBench-Psych-Hard 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
0.00 0.10 0.20 0.30 0.40 0.50 2024 2025 2026 0.00 0.10 0.20 0.30 0.40 0.50 2024 2025 2026

Source: Official HealthBench-Psych-Hard 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 Law, finance and medicine

All 637 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
  • Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (28 entries).
  • Pass Models scored 27 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from September 2026, for GPT-6.1 Sol.
  • Pass Frontier models 5 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 27 rows for models we track and 1 other entry 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