Benchmark Almanac

RareDiseaseBench

Gemini 4 Argon 43.8% leads since September 2026
Not in model cards

Agents combine clinical and biological evidence to diagnose and manage rare diseases. A rubric grades their diagnoses and plans.

LatchBio lists the evaluation in its V1 inventory.

Made by
LatchBio
Kind
Research science
Level
Expert
Scored
Checked automatically
Answers
Agent task
Measures
Capability
Language
English
In and out
Text → Actions, Text
Interaction
Agentic
Environment
APIs and tools
Run by
A host
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
October 2026
Scores refreshed
November 20th, 2026
20
Models scored on its own leaderboard

Gemini 4 Argon leads with 43.8%

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.

  1. 1 Gemini 4 Argon 43.8%
  2. 2 GPT-6 Astra 33.3%
  3. 3 Claude Opus 5.5 LSVP 33.3%
  4. 4 GPT-6.1 Sol 27.6%
  5. 5 GPT-6.1 Terra 25.7%
  6. 6 Claude Sonnet 5.5 LSVP 22.9%
  7. 7 Claude Fable 5.1 LSVP 19.1%
  8. 8 GPT-5.6 Sol 13.3%
  9. 9 Claude Opus 5 LSVP 11.4%
  10. 10 Grok 4.6 4.76%
  11. 11 Gemini 3.8 Flash 2.86%
  12. 12 Mistral Large 4 1.90%
  13. 13 Claude Sonnet 5 LSVP 0.95%
  14. 14 GPT-5.6 Luna 0.95%
  15. 15 Nemotron 3 Ultra 0%
  16. 16 Kimi K3 0%
  17. 17 Nemotron 3.5 Lightning 30B-A3B 0%
  18. 18 GLM-5.3 0%
  19. 19 DeepSeek V4.1 Flash 0%
  20. 20 GPT-6 Luna 0%

Source: Official RareDiseaseBench leaderboard

#ModelScoreSettingDateSource
1 Gemini 4 Argon 43.8% Pi Sep 2026 Official leaderboard
2 GPT-6 Astra 33.3% Pi Sep 2026 Official leaderboard
3 Claude Opus 5.5 33.3% LSVP, Claude Code Sep 2026 Official leaderboard
4 GPT-6.1 Sol 27.6% Pi Sep 2026 Official leaderboard
5 GPT-6.1 Terra 25.7% Pi Oct 2026 Official leaderboard
6 Claude Sonnet 5.5 22.9% LSVP, Claude Code Sep 2026 Official leaderboard
7 Claude Fable 5.1 19.1% LSVP, Claude Code Sep 2026 Official leaderboard
8 GPT-5.6 Sol 13.3% Pi Jul 2026 Official leaderboard
9 Claude Opus 5 11.4% LSVP, Claude Code Jul 2026 Official leaderboard
10 Grok 4.6 4.76% Pi Aug 2026 Official leaderboard
11 Gemini 3.8 Flash 2.86% Pi Sep 2026 Official leaderboard
12 Mistral Large 4 1.90% Pi Oct 2026 Official leaderboard
13 Claude Sonnet 5 0.95% LSVP, Claude Code Jun 2026 Official leaderboard
14 GPT-5.6 Luna 0.95% Pi Jul 2026 Official leaderboard
15 Nemotron 3 Ultra 0% Pi Jun 2026 Official leaderboard
16 Kimi K3 0% Pi Jul 2026 Official leaderboard
17 Nemotron 3.5 Lightning 30B-A3B 0% Pi Aug 2026 Official leaderboard
18 GLM-5.3 0% Pi Aug 2026 Official leaderboard
19 DeepSeek V4.1 Flash 0% Pi Sep 2026 Official leaderboard
20 GPT-6 Luna 0% Codex Sep 2026 Official leaderboard

6 models have held number one on the official leaderboard since 2023

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

NVIDIA Anthropic OpenAI Google DeepMind
Own leaderboard 6 leaders

Source: Official RareDiseaseBench 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 Best so far
0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Gemini 4 Argon GPT-6 Astra Claude Fable 5.1 GPT-5.6 Sol Claude Sonnet 5 0% 20% 40% 60% 80% 100% 2026 2027 Near the ceiling (85%) Gemini 4 Argon GPT-6 Astra

Source: Official RareDiseaseBench 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 Science and reasoning

All 203 in Science and reasoning
  • GPQA Diamond In use 198 graduate-level 'Google-proof' multiple-choice questions in biology, chemistry and physics. Gemini 4 Carbon 97%
  • Terminal-Bench-Science In use 70 scientific research-workflow tasks across five science areas, run in Harbor containers. Hidden tests give the resolution rate. GPT-6 Astra 68.1%
  • BioMysteryBench In use Bioinformatics problems on anonymized real datasets that require real analysis with tools. Rubric-graded, with human-solvable and human-difficult splits. Claude Fable 5.2 83.7%
  • FrontierScience In use Models solve physics, chemistry, and biology problems in an Olympiad track graded by short answer and a Research track graded by 10-point rubrics. Intern-S2-397B 87%
  • HiPhO In use Models solve theory problems from 2024 and 2025 physics olympiad exams, some with diagrams. Official marking schemes grade answers and steps, mapped to medal levels. Gemini 3 Pro 78.3%
  • LABBench2 In use 11 biology research sub-tasks (literature, figures, tables, patents, trials and more) with tool access. Macro-averaged accuracy. GPT-5.2 60.9%

Data audit

10 of 11 checks pass, 1 warning. 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 5 pass
  • Pass Independent runs Scores from the official leaderboard (20 entries).
  • Pass Models scored 20 models we track.
  • Pass Newest score The newest is from October 2026, for GPT-6.1 Terra.
  • Pass Frontier models 7 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Pass Headroom The best is 43.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on benchmarks.bio: 31 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