Benchmark Almanac

LiveTradeBench

Claude 3.7 Sonnet 12.1% leads since August 2025
Not in model cards

Agents allocate simulated portfolios across US securities and Polymarket contracts using streamed prices and news. The researchers track portfolio returns over successive trading days.

The authors evaluated 21 language models over 50 trading days. Their public evaluation code has 169 GitHub stars.

Released
August 2025
Kind
Finance
Level
University
Scored
Settled by events
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Time series → Actions
Interaction
Agentic
Items from
Live play or votes
Refresh
Live
Run by
Anyone
Strongest model tested
GPT-5 #71 on the capability index
Last model added
September 2025
Scores refreshed
November 20th, 2026
21
Entries on its own leaderboard, 18 models we track

Claude 3.7 Sonnet leads with 12.1% among models released by September 2025 Claude 3.7 Sonnet leads with 12.1% among models released by September 2025

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. This view shows only models we track, each at its best.

The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude 3.7 Sonnet 12.1%
  2. 2 Grok 4 5.84%
  3. 3 Qwen2.5-72B 3.39%
  4. 4 Llama 3.3 70B Instruct 2.15%
  5. 5 Claude Opus 4 0.94%
  6. 6 DeepSeek-V3.1 -1.11%
  7. 7 Grok 3 -2.56%
  8. 8 DeepSeek-R1 (snapshot unspecified) -5.54%
  9. 9 Llama 4 Scout -5.70%
  10. 10 Llama 4 Maverick -6.92%
  11. 11 GPT-5 -9.32%
  12. 12 Gemini 2.5 Flash Sep 2025 -10.15%
  13. 13 Claude Opus 4.1 -10.98%
  14. 14 GPT-4o (snapshot unspecified) -13.70%
  15. 15 GPT-4.1 -13.72%
  16. 16 Gemini-2.5-Pro (snapshot unspecified) -16.60%
  17. 17 Claude Sonnet 4 -18.93%
  18. 18 o3 -24.40%
  19. 19 Kimi K2 Jul 2025 -25.18%
  20. 20 Qwen3-235B-A22B-Instruct Jul 2025 -25.36%
  21. 21 Qwen3-235B-A22B-Thinking Jul 2025 -27.92%

Source: Official LiveTradeBench leaderboard

#EntryScoreSettingDateSource
1 Claude 3.7 Sonnet 12.1% Feb 2025 Official leaderboard
2 Grok 4 5.84% Jul 2025 Official leaderboard
3 Qwen2.5-72B 3.39% Sep 2024 Official leaderboard
4 Llama 3.3 70B Instruct 2.15% Dec 2024 Official leaderboard
5 Claude Opus 4 0.94% May 2025 Official leaderboard
6 DeepSeek-V3.1 -1.11% Aug 2025 Official leaderboard
7 Grok 3 -2.56% Apr 2025 Official leaderboard
8 DeepSeek-R1 (snapshot unspecified) -5.54% Official leaderboard
9 Llama 4 Scout -5.70% Apr 2025 Official leaderboard
10 Llama 4 Maverick -6.92% Apr 2025 Official leaderboard
11 GPT-5 -9.32% Aug 2025 Official leaderboard
12 Gemini 2.5 Flash Sep 2025 -10.15% Sep 2025 Official leaderboard
13 Claude Opus 4.1 -10.98% Aug 2025 Official leaderboard
14 GPT-4o (snapshot unspecified) -13.70% Official leaderboard
15 GPT-4.1 -13.72% Apr 2025 Official leaderboard
16 Gemini-2.5-Pro (snapshot unspecified) -16.60% Official leaderboard
17 Claude Sonnet 4 -18.93% May 2025 Official leaderboard
18 o3 -24.40% Apr 2025 Official leaderboard
19 Kimi K2 Jul 2025 -25.18% Jul 2025 Official leaderboard
20 Qwen3-235B-A22B-Instruct Jul 2025 -25.36% Jul 2025 Official leaderboard
21 Qwen3-235B-A22B-Thinking Jul 2025 -27.92% Jul 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Claude 3.7 Sonnet 12.1% Feb 2025 Official leaderboard
2 Grok 4 5.84% Jul 2025 Official leaderboard
3 Qwen2.5-72B 3.39% Sep 2024 Official leaderboard
4 Llama 3.3 70B Instruct 2.15% Dec 2024 Official leaderboard
5 Claude Opus 4 0.94% May 2025 Official leaderboard
6 DeepSeek-V3.1 -1.11% Aug 2025 Official leaderboard
7 Grok 3 -2.56% Apr 2025 Official leaderboard
8 Llama 4 Scout -5.70% Apr 2025 Official leaderboard
9 Llama 4 Maverick -6.92% Apr 2025 Official leaderboard
10 GPT-5 -9.32% Aug 2025 Official leaderboard
11 Gemini 2.5 Flash Sep 2025 -10.15% Sep 2025 Official leaderboard
12 Claude Opus 4.1 -10.98% Aug 2025 Official leaderboard
13 GPT-4.1 -13.72% Apr 2025 Official leaderboard
14 Claude Sonnet 4 -18.93% May 2025 Official leaderboard
15 o3 -24.40% Apr 2025 Official leaderboard
16 Kimi K2 Jul 2025 -25.18% Jul 2025 Official leaderboard
17 Qwen3-235B-A22B-Instruct Jul 2025 -25.36% Jul 2025 Official leaderboard
18 Qwen3-235B-A22B-Thinking Jul 2025 -27.92% Jul 2025 Official leaderboard

Rankings by part

We read 2 parts from the benchmark’s own leaderboard: U.S. Stock Market and Polymarket Prediction Market. Each ranks models on one part of the test. GPT-4.1 leads U.S. Stock Market and Claude 3.7 Sonnet leads Polymarket Prediction Market. Arrows show how far a model moved from its overall rank among the models we track.

U.S. Stock Market

18 models we track have a score here.

  1. 1 GPT-4.1 ↑12 6.25%
  2. 2 o3 ↑13 6.04%
  3. 3 GPT-5 ↑7 5.31%
  4. 4 Qwen2.5-72B ↓1 5.15%
  5. 5 Llama 4 Scout ↑3 4.65%
  6. 6 Llama 4 Maverick ↑3 4.46%
  7. 7 Grok 4 ↓5 4.30%
  8. 8 Claude Opus 4 ↓3 3.93%
  9. 9 Claude Opus 4.1 ↑3 3.73%
  10. 10 Claude 3.7 Sonnet ↓9 3.63%

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 Sep 2025

Source: Official LiveTradeBench 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% 2025 Near the ceiling (85%) Claude 3.7 Sonnet Qwen2.5-72B 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Claude 3.7 Sonnet Qwen2.5-72B

Source: Official LiveTradeBench 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 Newest score check. The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (21 entries).
  • Pass Models scored 18 models we track, plus 3 other entries on its leaderboard.
  • Warning Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5 (#71).
  • Pass Headroom The best is 12.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 18 rows for models we track and 3 other entries 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