Benchmark Almanac

Accounting Autocoding

Gemini 3.1 Pro 92.6% leads since March 2026
Not in model cards

Models select a bookkeeping category for each expense. Ramp checks the first choice against the assigned category.

Made by
Ramp Labs
Released
March 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
Refresh
One fixed set
Run by
A host
In views
Law and finance
Strongest model tested
GPT-5.4 #28 on the capability index
Last model added
March 2026
Scores refreshed
November 20th, 2026
21
Models scored on its own leaderboard

Gemini 3.1 Pro leads with 92.6% among models released by March 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 March 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 3.1 Pro 92.6%
  2. 2 Gemini 2.5 Pro Jun 2025 92%
  3. 3 Gemini 3 Pro 92%
  4. 4 Claude Opus 4.6 91.9%
  5. 5 GLM-5 91.6%
  6. 6 Claude Opus 4.5 91.5%
  7. 7 Gemini 3 Flash 91.4%
  8. 8 Claude Sonnet 4.6 91.3%
  9. 9 GPT-5.4 90.7%
  10. 10 Kimi K2.5 90.6%
  11. 11 Gemini 2.5 Flash Sep 2025 90.5%
  12. 12 Claude Sonnet 4.5 90.4%
  13. 13 Gemini 2.0 Flash Feb 2025 90.1%
  14. 14 GPT-5.1 90.1%
  15. 15 GPT-5 mini 90%
  16. 16 MiniMax-M2.5 90%
  17. 17 Claude Haiku 4.5 89.7%
  18. 18 GPT-5.2 89.4%
  19. 19 Gemini 2.0 Flash-Lite 88.5%
  20. 20 GPT-4.1 mini 88.2%
  21. 21 Gemini 2.5 Flash-Lite Sep 2025 87.9%

Source: Official Accounting Autocoding leaderboard

#ModelScoreSettingDateSource
1 Gemini 3.1 Pro 92.6% Feb 2026 Official leaderboard
2 Gemini 2.5 Pro Jun 2025 92% Jun 2025 Official leaderboard
3 Gemini 3 Pro 92% Nov 2025 Official leaderboard
4 Claude Opus 4.6 91.9% Feb 2026 Official leaderboard
5 GLM-5 91.6% Feb 2026 Official leaderboard
6 Claude Opus 4.5 91.5% Nov 2025 Official leaderboard
7 Gemini 3 Flash 91.4% Dec 2025 Official leaderboard
8 Claude Sonnet 4.6 91.3% Feb 2026 Official leaderboard
9 GPT-5.4 90.7% Mar 2026 Official leaderboard
10 Kimi K2.5 90.6% Jan 2026 Official leaderboard
11 Gemini 2.5 Flash Sep 2025 90.5% Sep 2025 Official leaderboard
12 Claude Sonnet 4.5 90.4% Sep 2025 Official leaderboard
13 Gemini 2.0 Flash Feb 2025 90.1% Feb 2025 Official leaderboard
14 GPT-5.1 90.1% Nov 2025 Official leaderboard
15 GPT-5 mini 90% Aug 2025 Official leaderboard
16 MiniMax-M2.5 90% Feb 2026 Official leaderboard
17 Claude Haiku 4.5 89.7% Oct 2025 Official leaderboard
18 GPT-5.2 89.4% Dec 2025 Official leaderboard
19 Gemini 2.0 Flash-Lite 88.5% Feb 2025 Official leaderboard
20 GPT-4.1 mini 88.2% Apr 2025 Official leaderboard
21 Gemini 2.5 Flash-Lite Sep 2025 87.9% Sep 2025 Official leaderboard

Number one over time

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

Google DeepMind
Own leaderboard Passed 85% in Mar 2026, by Gemini 3.1 Pro

Source: Official Accounting Autocoding leaderboard

Gemini 2.0 Flash on the official leaderboard passed 85% in February 2025

Models released before it came out already scored that high. 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% 2025 2026 Near the ceiling (85%) Gemini 3.1 Pro Gemini 2.5 Pro Gemini 2.0 Flash 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Gemini 3.1 Pro Gemini 2.0 Flash

Source: Official Accounting Autocoding 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 (21 entries).
  • Pass Models scored 21 models we track.
  • Note Newest score The newest came 8 months ago, in March 2026, for GPT-5.4. That’s expected for a saturated test.
  • Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5.4 (#28).
  • Note Headroom Saturated. The best is 92.6%, past 85% since February 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on builders.ramp.com: 21 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