# AIME25 (Arcee) leaderboard

Mathematics benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelscale.dev/leaderboards/benchmarks/math-aime2025-arcee

AIME25 first-party comparison snapshot. Every model the catalog carries a published AIME25 (Arcee) value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Mathematics |
| Measure | Integer answers 000-999 |
| Tasks | 15 problems |
| Difficulty | High school olympiad level |
| Published by | [Trinity-Large-Thinking: Scaling an Open Source Frontier Agent](https://www.arcee.ai/blog/trinity-large-thinking) |

A display-only AIME25 reference from Arcee AI's Trinity-Large-Thinking launch chart.

## AIME25 (Arcee) ranking

6 models with a published AIME25 (Arcee) value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Integer answers 000-999 | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Claude Opus 4.6](https://modelscale.dev/models/claude-opus-4-6) | Anthropic | 99.8 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Kimi K2.5](https://modelscale.dev/models/kimi-k2-5) | Moonshot AI | 96.3 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Trinity-Large-Thinking](https://modelscale.dev/models/trinity-large-thinking) | Arcee AI | 96.3 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GLM-5](https://modelscale.dev/models/glm-5) | Z.AI | 93.3 | Observed 2026-11-21 · source benchlm:models |
| 5 | [MiniMax M2.7](https://modelscale.dev/models/minimax-m2-7) | MiniMax | 80.0 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Trinity-Large-Preview](https://modelscale.dev/models/trinity-large-preview) | Arcee AI | 24.0 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
