Benchmark Almanac

FinDendro

GPT-5 0.76 leads since November 2026
Not in model cards

Models recall dated US financial facts and identify the limits of their knowledge. Four prompt styles test value recall, self-knowledge, and historical timing.

The suite contains 2,226 facts spanning 1999 through 2026. The chance-controlled baseline is about 0.27, so scores require the prompt protocol.

Released
November 2026
Kind
Finance
Level
Expert
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
In views
Law and finance
Strongest model tested
GPT-5 #71 on the capability index
Last model added
February 2026
Scores refreshed
November 20th, 2026

In the authors’ words

Financial backtests of large language models (LLMs) are meaningful only if the model did not already know the outcomes it is asked to predict. Practitioners guard against this by trusting a vendor's stated knowledge cutoff, by asking the model what it knows, or by instructing it to answer as of a past date. To verify whether any of these safeguards can be trusted, we introduce FinDendro, a benchmark of 2,226 dated U.S. facts from 1999 to 2026, covering monthly macroeconomic series, month-end stock prices and Federal Reserve decisions. Each fact yields four prompts: a value question, Recall and Predict questions that pair each candidate value with a claim of knowing or guessing without a declared date, and a Predict question with a historical declared date. Controls for persistence and dominant outcomes keep chance accuracy near 0.27. Results show that for most frontier models, self-reports understate knowledge. GPT-5 answers 76% of controlled facts correctly but claims to know 38%, and answers 61% of the facts it says it does not know. Knowledge claims can persist beyond the measured knowledge boundary, by up to seven quarters, and a declared date removes most of the claims of prior knowledge while leaving accuracy unchanged. Backtests inside a model's training window therefore cannot be certified leak-free by self-report, instruction or stated cutoff. These findings motivate behavioral measurement of temporal knowledge boundaries, and FinDendro provides a benchmark and procedure for selecting backtest periods based on demonstrated factual knowledge.

Abstract of LLMs Don't Know What They Know: Benchmarking Temporal Self-Knowledge for Financial Backtesting (arXiv:2611.06808, CC0)
6
Entries in its paper, 5 models we track

GPT-5 leads with 0.76 among models released by February 2026 GPT-5 leads with 0.76 among models released by February 2026

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in February 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 GPT-5 0.76
  2. 2 DeepSeek-V3.2 0.54
  3. 3 GPT-4o (snapshot unspecified) 0.50
  4. 4 Llama 3.3 70B 0.43
  5. 5 Qwen3.5-9B 0.37
  6. 6 Qwen3-8B 0.36

Source: The FinDendro paper

#EntryScoreSettingDateSource
1 GPT-5 0.76 Aug 2025 Official leaderboard
2 DeepSeek-V3.2 0.54 Dec 2025 Official leaderboard
3 GPT-4o (snapshot unspecified) 0.50 Official leaderboard
4 Llama 3.3 70B 0.43 Dec 2024 Official leaderboard
5 Qwen3.5-9B 0.37 Feb 2026 Official leaderboard
6 Qwen3-8B 0.36 Apr 2025 Official leaderboard
#ModelScoreSettingDateSource
1 GPT-5 0.76 Aug 2025 Official leaderboard
2 DeepSeek-V3.2 0.54 Dec 2025 Official leaderboard
3 Llama 3.3 70B 0.43 Dec 2024 Official leaderboard
4 Qwen3.5-9B 0.37 Feb 2026 Official leaderboard
5 Qwen3-8B 0.36 Apr 2025 Official leaderboard

Rankings by part

We read 8 parts from the benchmark’s paper, each ranking models on one part of the test, such as Recall: claims knowledge, Recall: accuracy when claiming knowledge and Recall: accuracy when denying knowledge. GPT-5 leads 4 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Recall: claims knowledge

5 models we track have a score here.

  1. 1 Qwen3-8B ↑4 0.71
  2. 2 Qwen3.5-9B ↑2 0.69
  3. 3 Llama 3.3 70B 0.48
  4. 4 GPT-5 ↓3 0.38
  5. 5 DeepSeek-V3.2 ↓3 0.28

Number one over time

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

OpenAI
Benchmark paper 1 leader
GPT-5 0.76

Source: The FinDendro paper

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
In the benchmark’s paper
0.40 0.50 0.60 0.70 0.80 2025 2026 0.40 0.50 0.60 0.70 0.80 2025 2026

Source: Official FinDendro 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

8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 9 months ago, in February 2026, for Qwen3.5-9B. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
  • Pass Independent runs Scores from the benchmark’s paper (6 entries).
  • Pass Models scored 5 models we track, plus 1 other entry on its leaderboard.
  • Warning Newest score The newest came 9 months ago, in February 2026, for Qwen3.5-9B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 5 rows for models we track and 1 other entry 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