DTBench leaderboard
As of 15 Nov 2026, the top score on DTBench is 99.2% by Claude Fable 5.2 (Anthropic), ahead of Claude Opus 5.5 at 98.9%, across 238 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 9 Nov 2026). Decision problems where the model's answer preference is compared with decision-theoretic expert answers. DTBench is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 15 Nov 2026)
- Claude Fable 5.2 holds the top DTBench score at 99.2% (Epoch AI Benchmarking Hub, 9 Nov 2026).
- Claude Opus 5.5 is second on DTBench at 98.9% (Epoch AI Benchmarking Hub, 22 Sep 2026).
- DTBench: Decision-theory dilemmas scored against expert answers (Epoch AI Benchmarking Hub).
- 238 model configurations have a published DTBench result on BenchLeader (BenchLeader, 15 Nov 2026).
- DTBench does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Decision-theory dilemmas scored against expert answers.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Reasoning
- Index weight
- Reference only
- Models
- 238
- Data as of
- 15 Nov 2026
- Newest result
- 9 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
What the test looks like
Decision problems where the model's answer preference is compared with decision-theoretic expert answers.
How it is scored
Accuracy, as published by DTBench.
What to keep in mind
Philosophy-of-decision framing; a large model set.
- 1Claude Fable 5.299.2%
- 2Claude Fable 5.2 (xhigh)98.9%
- 3Claude Opus 5.5 (max)98.9%
- 4Claude Fable 5 (max)98.4%
- 5Claude Fable 5.2 (high)98.1%
- 6Claude Fable 5.2 (medium)98.1%
- 7Claude Fable 5.2 (low)98.1%
- 8Claude Opus 5 (high)97.9%
- 9Claude Opus 5 (xhigh)97.9%
- 10Claude Fable 5.197.6%
- 11Claude Opus 5 (max)97.6%
- 12Claude Fable 5.1 (xhigh)97.6%
- 13Gemini 4 Argon (low)97.6%
- 14GPT-6 Astra (xhigh)97.3%
- 15Claude Opus 5.5 (xhigh)97.3%
| # | ||||
|---|---|---|---|---|
| 1 | New | 99.2% | 65.4 | 2026-11-09 |
| 2 | New | 98.9% | 69.7 | 2026-11-09 |
| 3 | 98.9% | 68.4 | 2026-09-22 | |
| 4 | 98.4% | 65.1 | 2026-06-09 | |
| 5 | New | 98.1% | 68.4 | 2026-11-09 |
| 6 | New | 98.1% | 66.3 | 2026-11-09 |
| 7 | New | 98.1% | 65.3 | 2026-11-09 |
| 8 | 97.9% | 64.4 | 2026-07-24 | |
| 9 | 97.9% | 60.1 | 2026-07-24 | |
| 10 | 97.6% | 66.4 | 2026-09-01 | |
| 11 | 97.6% | 65.0 | 2026-07-24 | |
| 12 | 97.6% | 64.6 | 2026-09-01 | |
| 13 | 97.6% | – | 2026-10-13 | |
| 14 | 97.3% | 66.0 | 2026-09-03 | |
| 15 | 97.3% | 65.4 | 2026-09-22 | |
| 16 | 97.3% | 65.3 | 2026-10-13 | |
| 17 | 97.3% | 64.2 | 2026-09-28 | |
| 18 | 97.3% | 62.6 | 2026-09-22 | |
| 19 | 97.3% | 61.6 | 2026-10-13 | |
| 20 | 97.3% | 56.1 | 2026-08-12 | |
| 21 | 97.3% | – | 2026-06-09 | |
| 22 | 97.1% | 66.4 | 2026-09-03 | |
| 23 | 97.1% | 64.4 | 2026-09-03 | |
| 24 | 97.1% | 61.7 | 2026-09-01 | |
| 25 | 97.1% | – | 2026-02-19 | |
| 26 | 96.8% | 62.1 | 2026-08-13 | |
| 27 | 96.5% | 66.6 | 2026-09-29 | |
| 28 | 96.5% | 66.3 | 2026-09-03 | |
| 29 | 96.5% | 66.0 | 2026-09-22 | |
| 30 | 96.5% | 63.9 | 2026-06-09 | |
| 31 | 96.5% | 63.7 | 2026-09-01 | |
| 32 | 96.5% | 61.6 | 2026-09-02 | |
| 33 | 96.5% | 61.2 | 2026-09-22 | |
| 34 | 96.5% | 56.8 | 2026-07-08 | |
| 35 | 96.0% | 64.1 | 2026-04-23 | |
| 36 | 96.0% | 62.5 | 2026-09-28 | |
| 37 | 96.0% | 57.4 | 2026-09-21 | |
| 38 | 96.0% | – | 2026-04-23 | |
| 39 | 96.0% | – | 2026-07-09 | |
| 40 | 95.7% | 63.9 | 2026-07-09 | |
| 41 | 95.7% | – | 2026-06-09 | |
| 42 | 95.7% | 56.4 | 2026-09-02 | |
| 43 | 95.5% | 65.3 | 2026-07-09 | |
| 44 | 95.5% | 63.4 | 2026-07-09 | |
| 45 | 95.5% | 61.0 | 2026-07-21 | |
| 46 | 95.5% | 60.4 | 2026-02-19 | |
| 47 | 95.5% | 58.9 | 2026-09-22 | |
| 48 | 95.2% | 64.4 | 2026-10-12 | |
| 49 | 95.2% | 61.3 | 2026-07-09 | |
| 50 | 95.0% | 55.3 | 2026-10-07 | |
| 51 | 94.9% | 62.0 | 2026-09-03 | |
| 52 | 94.9% | 61.1 | 2026-05-28 | |
| 53 | 94.9% | 60.2 | 2026-09-02 | |
| 54 | 94.9% | 57.5 | 2026-07-24 | |
| 55 | 94.9% | 55.9 | 2026-09-02 | |
| 56 | 94.9% | – | 2026-02-19 | |
| 57 | 94.7% | 62.5 | 2026-05-19 | |
| 58 | 94.7% | 61.5 | 2026-08-05 | |
| 59 | 94.7% | 60.1 | 2026-08-13 | |
| 60 | 94.7% | 59.7 | 2026-04-16 |
AI reasoning and math benchmark leaderboard · All LLM benchmarks
Related benchmarks
Top head-to-heads on DTBench
Cite as: BenchLeader, “DTBench leaderboard”, data as of 15 Nov 2026, https://www.benchleader.com/benchmarks/dtbench.
DTBench: questions
- What does DTBench measure?
- Decision problems where the model's answer preference is compared with decision-theoretic expert answers. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on DTBench?
- As of 15 Nov 2026, the highest score on DTBench is 99.2% by Claude Fable 5.2 (result dated 9 Nov 2026), ahead of Claude Opus 5.5 at 98.9%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads DTBench?
- Claude Fable 5.2 leads DTBench with 99.2% as of 15 Nov 2026 (result dated 9 Nov 2026), ahead of Claude Opus 5.5 at 98.9%.
- How many models have DTBench results?
- 238 model configurations have a DTBench result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs DTBench and how often is it updated?
- DTBench is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does DTBench count toward the BenchLeader Index?
- No. DTBench is shown for reference but left out of the composite index.