# LiveTradeBench

Live portfolio allocation by LLMs on streamed prices and news, run in two different markets: US stocks and Polymarket prediction markets. Each step the model outputs percentage weights.

## Facts

- Maker: University of Illinois Urbana-Champaign (U Lab)
- Year: 2025
- Use case: Financial trading and portfolios
- Size: 21 LLMs; 50 trading days (2025-08-18 to 2025-10-24); 15 US equities/ETFs + cash, 10 Polymarket markets
- Public split: Code and PyPI package; live site
- Grading: Cumulative return and Sharpe ratio (plus drawdown/volatility) per market
- Licence: PolyForm Noncommercial 1.0.0 (code); separate commercial licence
- Submission: maker-only
- Products allowed: Not stated
- Latest version: arXiv v1 (2025-11-05)
- What it does not cover: Order execution and trading costs at realistic size; markets other than US stocks and Polymarket
- Last checked: 1 Nov 2026
- Links: paper https://arxiv.org/abs/2511.03628, site https://trade-bench.live, code https://github.com/ulab-uiuc/live-trade-bench
- Directory page: https://benchmarks.finance/benchmarks/livetradebench/
- Leaderboard: https://benchmarks.finance/leaderboards/livetradebench/

## Results

### LiveTradeBench: cumulative return % (higher is better, 10 models)

| Rank | Model | Provider | Score | Metric | Setting | Reported by | Source | URL | Date |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | GPT-4.1 | OpenAI | 6.3 | cumulative return % | US stock market, live 2025-08-18 to 2025-10-24 (50 trading days) | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 2 | GPT-o3 | OpenAI | 6.0 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 3 | GPT-5 | OpenAI | 5.3 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 4 | Qwen2.5-72B-Instruct | Qwen | 5.2 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 5 | Llama4-Scout | Meta | 4.7 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 6 | Llama4-Maverick | Meta | 4.5 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 7 | Grok 4 | xAI | 4.3 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 8 | Claude Opus 4 | Anthropic | 3.9 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 9 | Claude Opus 4.1 | Anthropic | 3.7 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |
| 10 | Claude-Sonnet-3.7 | Anthropic | 3.6 | cumulative return % | US stock market, live | Reported by the authors | Paper Table 2 | https://arxiv.org/html/2511.03628v1 | 5 Nov 2025 (checked 1 Nov 2026) |


Source: https://benchmarks.finance/benchmarks/livetradebench/

Created and maintained by Moneda.