Benchmark Almanac

Agentic Credit Analysis

Ramp pipeline + GPT-6 Luna 84.4% tops its leaderboard Claude Sonnet 5.5 80.2% leads as a model on its own
Not in model cards

Models extract and standardize income statements and balance sheets. Ramp compares their figures with credit analysts’ figures using a five percent tolerance.

Made by
Ramp Labs
Released
October 2026
Kind
Finance
Level
Expert
Scored
Checked automatically
Answers
Agent task
Tasks
Kept private
Measures
Capability
Language
English
In and out
Text, Documents → Tables, Actions
Interaction
Agentic
Environment
APIs and tools
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Occupations
Business and financial operations
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
September 2026
Scores refreshed
November 20th, 2026
14
Entries on its own leaderboard, 7 models we track

Ramp pipeline + GPT-6 Luna tops the leaderboard with 84.4% Claude Sonnet 5.5 leads with 80.2% 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 on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. The best model on its own is Claude Sonnet 5.5, at 80.2%. Agents built around a model, such as Ramp pipeline + GPT-6 Luna, are their own entries. We don't count them as the model's score. 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 Ramp pipeline + GPT-6 Luna 84.4%
  2. 2 Ramp pipeline + GPT-6 Sol 84%
  3. 3 Ramp pipeline + Claude Sonnet 5.5 83.3%
  4. 4 Ramp pipeline + Claude Opus 5.5 82%
  5. 5 Ramp pipeline + Gemini 3.1 Pro 80.3%
  6. 6 Claude Sonnet 5.5 80.2%
  7. 7 Claude Opus 5.5 80%
  8. 8 Gemini 3.8 Flash 78.5%
  9. 9 Gemini 3.1 Pro 78.3%
  10. 10 GPT-6 Sol 78.3%
  11. 11 GPT-6 Luna 77.9%
  12. 12 Ramp pipeline + Gemini 3.8 Flash 77.6%
  13. 13 Ramp pipeline + GPT-6 Astra 77.3%
  14. 14 GPT-6 Astra 61.1%

Source: Official Agentic Credit Analysis leaderboard

#EntryScoreSettingDateSource
1 Ramp pipeline + GPT-6 Luna 84.4% Sep 2026 Official leaderboard
2 Ramp pipeline + GPT-6 Sol 84% Sep 2026 Official leaderboard
3 Ramp pipeline + Claude Sonnet 5.5 83.3% Sep 2026 Official leaderboard
4 Ramp pipeline + Claude Opus 5.5 82% Sep 2026 Official leaderboard
5 Ramp pipeline + Gemini 3.1 Pro 80.3% Feb 2026 Official leaderboard
6 Claude Sonnet 5.5 80.2% Sep 2026 Official leaderboard
7 Claude Opus 5.5 80% Sep 2026 Official leaderboard
8 Gemini 3.8 Flash 78.5% Sep 2026 Official leaderboard
9 Gemini 3.1 Pro 78.3% Feb 2026 Official leaderboard
10 GPT-6 Sol 78.3% Sep 2026 Official leaderboard
11 GPT-6 Luna 77.9% Sep 2026 Official leaderboard
12 Ramp pipeline + Gemini 3.8 Flash 77.6% Sep 2026 Official leaderboard
13 Ramp pipeline + GPT-6 Astra 77.3% Sep 2026 Official leaderboard
14 GPT-6 Astra 61.1% Sep 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Sonnet 5.5 80.2% Sep 2026 Official leaderboard
2 Claude Opus 5.5 80% Sep 2026 Official leaderboard
3 Gemini 3.8 Flash 78.5% Sep 2026 Official leaderboard
4 Gemini 3.1 Pro 78.3% Feb 2026 Official leaderboard
5 GPT-6 Sol 78.3% Sep 2026 Official leaderboard
6 GPT-6 Luna 77.9% Sep 2026 Official leaderboard
7 GPT-6 Astra 61.1% Sep 2026 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s own leaderboard: P&L and Balance sheet. Each ranks models on one part of the test. Claude Opus 5.5 leads P&L and Gemini 3.1 Pro leads Balance sheet. Arrows show how far a model moved from its overall rank among the models we track.

P&L

7 models we track have a score here.

  1. 1 Claude Opus 5.5 ↑1 77.6%
  2. 2 Claude Sonnet 5.5 ↓1 77%
  3. 3 GPT-6 Sol ↑2 76.3%
  4. 4 GPT-6 Luna ↑2 75.7%
  5. 5 Gemini 3.8 Flash ↓2 74.2%
  6. 6 Gemini 3.1 Pro ↓2 73.2%
  7. 7 GPT-6 Astra 60.5%

Number one over time

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

Anthropic
Own leaderboard Not tested since Oct 2026

Source: Official Agentic Credit Analysis 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%) Claude Sonnet 5.5 Gemini 3.1 Pro 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Claude Sonnet 5.5 Gemini 3.1 Pro

Source: Official Agentic Credit Analysis 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 11 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (14 entries).
  • Pass Models scored 7 models we track, plus 7 other entries on its leaderboard.
  • Pass Newest score The newest is from September 2026, for Claude Sonnet 5.5.
  • Pass Frontier models 3 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Pass Headroom The best is 80.2%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on builders.ramp.com: 14 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