Mathematics · Benchmark profile
A November 2025 HMMT slice for high-end mathematical reasoning comparisons.
Data verified 21 Jul 2026 · Methodology 1.6.0
Benchmark score on Harvard-MIT Mathematics Tournament November 2025
GLM-5 leads at 96.9%, followed by Qwen3.6 Plus (94.6%) and GLM-5.2 (94.4%).
Visual analysis
Switch between model placement, score distribution and descriptive provider averages. Every view uses the same sourced leaderboard.
9 ranked models · higher is better
| Model | Rank | Provider | Score |
|---|---|---|---|
| GLM-5 | #1 | Z.AI | 96.9% |
| Qwen3.6 Plus | #2 | Alibaba Cloud | 94.6% |
| GLM-5.2 | #3 | Z.AI | 94.4% |
| GLM-5.1 | #4 | Z.AI | 94% |
| Claude Opus 4.5 | #5 | Anthropic | 93.3% |
| Qwen3.5 397B | #6 | Alibaba Cloud | 92.7% |
| Kimi K2.5 | #7 | Moonshot AI | 91.1% |
| Qwen3.6 27B | #8 | Alibaba Cloud | 90.7% |
| Qwen3.6-35B-A3B | #9 | Alibaba Cloud | 89.1% |
One best score per model · higher is better
| Rank | Model | Provider | License | Evidence use | Score |
|---|---|---|---|---|---|
| #1 | GLM-5 glm-5 | Z.AI | open | Reference only | 96.9% |
| #2 | Qwen3.6 Plus qwen3-6-plus | Alibaba Cloud | closed | Reference only | 94.6% |
| #3 | GLM-5.2 glm-5-2 | Z.AI | open | Estimated reference | 94.4% |
| #4 | GLM-5.1 glm-5-1 | Z.AI | open | Reference only | 94% |
| #5 | Claude Opus 4.5 claude-opus-4-5 | Anthropic | closed | Reference only | 93.3% |
| #6 | Qwen3.5 397B qwen3.5-397b-a17b | Alibaba Cloud | open | Estimated reference | 92.7% |
| #7 | Kimi K2.5 kimi-k2.5 | Moonshot AI | — | Reference only | 91.1% |
| #8 | Qwen3.6 27B qwen3.6-27b | Alibaba Cloud | open | Estimated reference | 90.7% |
| #9 | Qwen3.6-35B-A3B qwen3-6-35b-a3b | Alibaba Cloud | open | Estimated reference | 89.1% |
The top of this snapshot is led by GLM-5 at 96.9%; third place is 2.5 points behind. The top-9 spread is 7.8 points.
About Harvard-MIT Mathematics Tournament November 2025
A November 2025 HMMT slice for high-end mathematical reasoning comparisons. Results stay tied to the exact model variant and evaluation system. Multiple systems for the same model use the best published score on this page; overall Lumina scoring uses the median of ranking-eligible rows.
Open benchmark source ↗FAQ
A November 2025 HMMT slice for high-end mathematical reasoning comparisons.
GLM-5 by Z.AI currently leads with 96.9%.
9 models in the LuminaBench cohort have a qualifying score on this benchmark.
No. This benchmark is display-only and does not enter the overall Lumina composite.
Related
