Wikibench

HMMT 2025

math official site →

Multilingual: no

Harvard-MIT Mathematics Tournament 2025 - A prestigious student-organized mathematics competition for high school students featuring two tournaments (November 2025 at MIT and February 2026 at Harvard) with individual tests, team rounds, and guts rounds

Methodology

Imported from llm-stats public benchmark metadata. Modality: text. Max score: 1. Categories: math. Language: en. Verified by llm-stats: no.

Leaderboard

Showing 20 of 29 results

  1. GPT-5.2 Pro self-reported llm-stats
    100.0%
  2. GPT-5.2 self-reported llm-stats
    99.4%
  3. DeepSeek-V3.2-Speciale self-reported llm-stats
    99.2%
  4. Kimi K2-Thinking-0905 self-reported llm-stats
    97.5%
  5. Qwen3.6 Plus self-reported llm-stats
    96.7%
  6. Kimi K2.5 self-reported llm-stats
    95.4%
  7. Qwen3.5-397B-A17B self-reported llm-stats
    94.8%
  8. Nemotron 3 Super (120B A12B) self-reported llm-stats
    94.7%
  9. GLM-5.1 self-reported llm-stats
    94.0%
  10. Qwen3.6-27B self-reported llm-stats
    93.8%
  11. GPT-5 self-reported llm-stats
    93.3%
  12. Grok 4 Fast self-reported llm-stats
    93.3%
  13. Qwen3.5-27B self-reported llm-stats
    92.0%
  14. Qwen3.5-122B-A10B self-reported llm-stats
    91.4%
  15. DeepSeek-V3.2 (Thinking) self-reported llm-stats
    90.2%
  16. DeepSeek-V3.2 self-reported llm-stats
    90.2%
  17. Qwen3.5-35B-A3B self-reported llm-stats
    89.0%
  18. GPT-5 mini self-reported llm-stats
    87.8%
  19. Sarvam-105B self-reported llm-stats
    85.8%
  20. MiMo-V2-Flash self-reported llm-stats
    84.4%