Benchmark Almanac

Agentic Trader

MAESTRO Multi Agent 9323.7% tops its leaderboard Gemini 3.1 Pro 79.4% leads the models we track
Not in model cards

Agents search for market information and place simulated foreign exchange orders using live prices. Each agent manages its own positions, cash, and margin.

HKU reported six weeks of trading by ten models. Its public leaderboard includes a separate multi-agent entry with a shorter trading history.

Released
July 2026
Kind
Finance
Level
University
Scored
Settled by events
Answers
Agent task
Measures
Capability
Language
English
In and out
Text, Time series → Actions
Interaction
Agentic
Refresh
Live
Run by
A host
Strongest model tested
Claude Opus 5 #11 on the capability index
Last model added
July 2026
Scores refreshed
November 20th, 2026
11
Entries on its own leaderboard, 10 models we track

MAESTRO Multi Agent tops the leaderboard with 9323.7% Gemini 3.1 Pro leads with 79.4% among models released by July 2026

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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. The best model we track is Gemini 3.1 Pro, at 79.4%. This view shows only models we track, each at its best.

The newest model with a score here came out in July 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 MAESTRO Multi Agent 9323.7%
  2. 2 Gemini 3.1 Pro 79.4%
  3. 3 DeepSeek-V4-Pro 36.7%
  4. 4 MiniMax-M3 11.4%
  5. 5 Seed2.0 Lite (0215) 4.48%
  6. 6 Claude Opus 5 -2.18%
  7. 7 GLM-5.2 -4.94%
  8. 8 GPT-5.6 Sol -6.51%
  9. 9 Qwen3.7-Max -10.08%
  10. 10 Kimi K3 -14.39%
  11. 11 Grok 4.5 -27.54%

Source: Official Agentic Trader leaderboard

#EntryScoreSettingDateLinkSource
1 MAESTRO Multi Agent 9323.7% maestro.trade-agent-aiel-hku.com Official leaderboard
2 Gemini 3.1 Pro 79.4% Feb 2026 Official leaderboard
3 DeepSeek-V4-Pro 36.7% Apr 2026 Official leaderboard
4 MiniMax-M3 11.4% Jun 2026 Official leaderboard
5 Seed2.0 Lite (0215) 4.48% Feb 2026 Official leaderboard
6 Claude Opus 5 -2.18% Jul 2026 Official leaderboard
7 GLM-5.2 -4.94% Jun 2026 Official leaderboard
8 GPT-5.6 Sol -6.51% Jul 2026 Official leaderboard
9 Qwen3.7-Max -10.08% May 2026 Official leaderboard
10 Kimi K3 -14.39% Jul 2026 Official leaderboard
11 Grok 4.5 -27.54% Jul 2026 Official leaderboard
#ModelScoreSettingDateLinkSource
1 Gemini 3.1 Pro 79.4% Feb 2026 Official leaderboard
2 DeepSeek-V4-Pro 36.7% Apr 2026 Official leaderboard
3 MiniMax-M3 11.4% Jun 2026 Official leaderboard
4 Seed2.0 Lite (0215) 4.48% Feb 2026 Official leaderboard
5 Claude Opus 5 -2.18% Jul 2026 Official leaderboard
6 GLM-5.2 -4.94% Jun 2026 Official leaderboard
7 GPT-5.6 Sol -6.51% Jul 2026 Official leaderboard
8 Qwen3.7-Max -10.08% May 2026 Official leaderboard
9 Kimi K3 -14.39% Jul 2026 Official leaderboard
10 Grok 4.5 -27.54% Jul 2026 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 Not tested since Jul 2026

Source: Official Agentic Trader 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 3.1 Pro Seed2.0 Lite (0215) 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro Seed2.0 Lite (0215)

Source: Official Agentic Trader 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 (11 entries).
  • Pass Models scored 10 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from July 2026, for Claude Opus 5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Opus 5 (#11).
  • Pass Headroom The best is 79.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on trade-agent-aiel-hku.com: 10 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