Benchmark Almanac

InsureBench Decision Models (Broader Comparison)

Gemini Decision 1 98% tops its leaderboard
Not in model cards

Models read text evidence and return structured findings about insurance scenarios. The scorer compares 150 findings from 50 packets with the author's labels.

The packets cover 25 related scenarios. InsureThing excludes 50 diagnostic findings from accuracy and reports its 100-question timing runs separately.

Made by
InsureThing
Released
October 2026
Kind
Finance
Level
University
Scored
Checked automatically
Answers
Multiple choice
Tasks
Kept private
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Written by experts
Refresh
One fixed set
Run by
A host
In views
Law and finance
33
Entries on its own leaderboard, 0 models we track

Gemini Decision 1 tops the leaderboard with 98%

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.

  1. 1 Gemini Decision 1 API 98%
  2. 2 Liquid d1-ultra API 96.7%
  3. 3 Perplexity Decider v1.2 · Oct16 API 96.7%
  4. 4 JEV 1.14 API 96%
  5. 5 Perplexity Decider v1.2 API 96%
  6. 6 Drex Max API 95.3%
  7. 7 Liquid d1.1 API 95.3%
  8. 8 Mercury Decide · Free API 94.7%
  9. 9 Mammoth 1 (free) API 94.7%
  10. 10 Mammoth 1 API 94.7%
  11. 11 Morph Reflex API 94.7%
  12. 12 Perplexity Decider v1 API 94%
  13. 13 Perplexity Decider v1.1 API 94%
  14. 14 Elephant 1 (free) API 93.3%
  15. 15 Elephant 1 API 93.3%
  16. 16 Luna Decisions API 92.7%
  17. 17 Mercury Decide API 92.7%
  18. 18 Parallel Decisions Fast API 92.7%
  19. 19 JEV 1.13 API 92%
  20. 20 Microsoft Decision-1 API 92%
  21. 21 Drex v2.0 API 92%
  22. 22 Perplexity Decider Flash v1.2 · Oct16 API 91.3%
  23. 23 Clef API 90%
  24. 24 Solar Decide API 89.3%
  25. 25 Liquid d1 API 88.7%
  26. 26 Perplexity Decider Flash v1.2 API 88%
  27. 27 Drex v1.5 API 86%
  28. 28 Clef Flash API 85.3%
  29. 29 KEV 4B API 84.7%
  30. 30 Span-02 API 84%
  31. 31 TEV 4B Experimental API 79.3%
  32. 32 Clef Omni API 76%
  33. 33 Joey 1 API 72%

Source: Official InsureBench Decision Models (Broader Comparison) leaderboard

#EntryScoreSettingDateSource
1 Gemini Decision 1 closed weights 98% API Nov 2026 Official leaderboard
2 Liquid d1-ultra closed weights 96.7% API Oct 2026 Official leaderboard
3 Perplexity Decider v1.2 · Oct16 closed weights 96.7% API Oct 2026 Official leaderboard
4 JEV 1.14 closed weights 96% API Oct 2026 Official leaderboard
5 Perplexity Decider v1.2 closed weights 96% API Oct 2026 Official leaderboard
6 Drex Max closed weights 95.3% API Oct 2026 Official leaderboard
7 Liquid d1.1 closed weights 95.3% API Oct 2026 Official leaderboard
8 Mercury Decide · Free closed weights 94.7% API Official leaderboard
9 Mammoth 1 (free) closed weights 94.7% API Oct 2026 Official leaderboard
10 Mammoth 1 closed weights 94.7% API Oct 2026 Official leaderboard
11 Morph Reflex closed weights 94.7% API Nov 2026 Official leaderboard
12 Perplexity Decider v1 closed weights 94% API Official leaderboard
13 Perplexity Decider v1.1 closed weights 94% API Official leaderboard
14 Elephant 1 (free) closed weights 93.3% API Oct 2026 Official leaderboard
15 Elephant 1 closed weights 93.3% API Oct 2026 Official leaderboard
16 Luna Decisions closed weights 92.7% API Official leaderboard
17 Mercury Decide closed weights 92.7% API Oct 2026 Official leaderboard
18 Parallel Decisions Fast closed weights 92.7% API Nov 2026 Official leaderboard
19 JEV 1.13 closed weights 92% API Official leaderboard
20 Microsoft Decision-1 closed weights 92% API Oct 2026 Official leaderboard
21 Drex v2.0 closed weights 92% API Oct 2026 Official leaderboard
22 Perplexity Decider Flash v1.2 · Oct16 closed weights 91.3% API Oct 2026 Official leaderboard
23 Clef closed weights 90% API Official leaderboard
24 Solar Decide closed weights 89.3% API Official leaderboard
25 Liquid d1 closed weights 88.7% API Official leaderboard
26 Perplexity Decider Flash v1.2 closed weights 88% API Oct 2026 Official leaderboard
27 Drex v1.5 open weights 86% API Oct 2026 Official leaderboard
28 Clef Flash closed weights 85.3% API Official leaderboard
29 KEV 4B open weights 84.7% API Official leaderboard
30 Span-02 open weights 84% API Oct 2026 Official leaderboard
31 TEV 4B Experimental open weights 79.3% API Official leaderboard
32 Clef Omni open weights 76% API Oct 2026 Official leaderboard
33 Joey 1 closed weights 72% API Oct 2026 Official 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 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the official leaderboard (33 entries).
  • Pass Models scored 33 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom No scores yet.
Leaderboard import 3 pass
  • Pass Importer We read the board on insurebench.insure-thing.com: 33 entries at the last read, none a model we track.
  • 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