GPT ranks 9th, DeepSeek surpasses Claude, the most powerful trading agent is actually the "outdated" 4.6

Deep Tide TechFlow News, on May 13th, the AI Trading Agent Benchmark, initiated by Manic.Trade, officially concluded. A total of 387 evaluations were completed, covering 180 testers, 46 models, and over 15 model families, making it the most comprehensive cross-model evaluation of trading agents to date. The assessment mainly focused on five dimensions: real-time data processing, multi-source intelligence gathering, market analysis, trading decision-making, and risk control.

The results show that the Claude family ranked first with 75.9 points, DeepSeek came in second with 74.4 points, and the GPT family was only ninth. In terms of individual models, Claude 4.7 Opus achieved the highest average score of 80.8; DeepSeek V4 Pro ranked second with 78.6; Claude 4.6 Opus scored 78.5 and ranked third, also delivering the only S-grade performance in the evaluation, with a top score of 90/100.

Notably, DeepSeek scored 15.9 in the Market Analysis dimension, surpassing Claude’s 15.4. Manic.Trade stated that this evaluation indicates that the current gap among AI trading agents is not mainly in their ability to make trading decisions, but rather in their capabilities for multi-source information integration, verifiable data processing, and risk management—functions closer to real trading execution.

Currently, users can visit Manic.Trade to view the full report and experience a crypto price prediction market designed for human traders and AI agents.

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin