Benchmark Almanac

Realm Financial Reasoning

Gemini 3.8 Flash 56.2% leads since September 2026
Not in model cards

Agents analyze financial documents and spreadsheets. A model grades their answers against practitioner rubrics.

Made by
micro1
Kind
Finance
Level
Expert
Scored
Graded by a model
Answers
Agent task
Measures
Capability
Language
English
In and out
Text, Documents, Tables → Text, Actions
Interaction
Agentic
Environment
Terminal, Code sandbox
Refresh
One fixed set
Run by
A host
Occupations
Business and financial operations
Strongest model tested
Claude Fable 5.1 #8 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
17
Models scored on its own leaderboard

Gemini 3.8 Flash leads with 56.2% among models released by September 2026

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.

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 Gemini 3.8 Flash high 56.2%
  2. 2 Muse Spark 1.1 xhigh 53.7%
  3. 3 Claude Opus 5 max 51.5%
  4. 4 Claude Fable 5 51.2%
  5. 5 Kimi K3 max 51.1%
  6. 6 Claude Fable 5.1 max 51.1%
  7. 7 GPT-5.5 50.8%
  8. 8 Muse Spark 1.3 xhigh 50.4%
  9. 9 Claude Sonnet 5 50%
  10. 10 Claude Opus 4.8 49.5%
  11. 11 Grok 4.6 high 49.5%
  12. 12 Grok 4.7 xhigh 49.5%
  13. 13 Grok 4.5 high 48.6%
  14. 14 Claude Opus 4.7 48.4%
  15. 15 GPT-5.6 Sol max 47.9%
  16. 16 Gemini 3.1 Pro 46.8%
  17. 17 Inkling medium 43.6%

Source: Official Realm Financial Reasoning leaderboard

#ModelScoreSettingDateSource
1 Gemini 3.8 Flash 56.2% high Sep 2026 Official leaderboard
2 Muse Spark 1.1 53.7% xhigh Jul 2026 Official leaderboard
3 Claude Opus 5 51.5% max Jul 2026 Official leaderboard
4 Claude Fable 5 51.2% Jun 2026 Official leaderboard
5 Kimi K3 51.1% max Jul 2026 Official leaderboard
6 Claude Fable 5.1 51.1% max Sep 2026 Official leaderboard
7 GPT-5.5 50.8% Apr 2026 Official leaderboard
8 Muse Spark 1.3 50.4% xhigh Sep 2026 Official leaderboard
9 Claude Sonnet 5 50% Jun 2026 Official leaderboard
10 Claude Opus 4.8 49.5% May 2026 Official leaderboard
11 Grok 4.6 49.5% high Aug 2026 Official leaderboard
12 Grok 4.7 49.5% xhigh Sep 2026 Official leaderboard
13 Grok 4.5 48.6% high Jul 2026 Official leaderboard
14 Claude Opus 4.7 48.4% Apr 2026 Official leaderboard
15 GPT-5.6 Sol 47.9% max Jul 2026 Official leaderboard
16 Gemini 3.1 Pro 46.8% Feb 2026 Official leaderboard
17 Inkling 43.6% medium Jul 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.

Google DeepMind Anthropic OpenAI Meta
Own leaderboard Not tested since Sep 2026

Source: Official Realm Financial Reasoning 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 Near the ceiling (85%) Gemini 3.8 Flash Claude Fable 5 Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.8 Flash Claude Opus 4.7

Source: Official Realm Financial Reasoning 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

9 of 10 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 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (17 entries).
  • Pass Models scored 17 models we track.
  • Pass Newest score The newest is from September 2026, for Grok 4.7.
  • Note Frontier models Only 1 of the top 10 has a score. The strongest tested is Claude Fable 5.1 (#8).
  • Pass Headroom The best is 56.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on micro1.ai: 17 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