← All visual benchmarksArena / 2026-10-22
Arena · REASONING

Mathematical reasoning

Human preference on mathematics prompts.

Hover or tap a bar to inspect. Select up to 3 models to highlight and compare.
Model scores10 models
Mathematical reasoning. Zoomed horizontal axis, from 1505.2 to 1542.4ZOOMED X-AXIS · 1505.2 – 1542.4 · NOT FROM ZERO1505.21514.51523.81533.11542.4Gemini 4 Argon (High)Gemini 4 Argon (High) · 1537.8 Arena Rating · zoomed horizontal axis1537.8Claude Opus 5 (Max)Claude Opus 5 (Max) · 1533 Arena Rating · zoomed horizontal axis1533Claude Fable 5.1 (Max)Claude Fable 5.1 (Max) · 1532.3 Arena Rating · zoomed horizontal axis1532.3Gemini 3.8 Flash (High)Gemini 3.8 Flash (High) · 1531.6 Arena Rating · zoomed horizontal axis1531.6Claude Opus 5 (High)Claude Opus 5 (High) · 1530.8 Arena Rating · zoomed horizontal axis1530.8Claude Opus 5.5 (High)Claude Opus 5.5 (High) · 1517.8 Arena Rating · zoomed horizontal axis1517.8Claude Opus 4.6 (High)Claude Opus 4.6 (High) · 1517.7 Arena Rating · zoomed horizontal axis1517.7Claude Fable 5 (High)Claude Fable 5 (High) · 1516.3 Arena Rating · zoomed horizontal axis1516.3Claude Opus 4.6Claude Opus 4.6 · 1510.7 Arena Rating · zoomed horizontal axis1510.7Gemini 3.7 Flash (High)Gemini 3.7 Flash (High) · 1509.8 Arena Rating · zoomed horizontal axis1509.8
Choose a bar for details.

How to read this chart

Official top-ranked excerpt. Human preference ratings are not accuracy percentages; IPS is a distinct metric.

Score unit: Arena Rating. Values can only be compared within the same task and protocol.

Source publication: 2026-10-22. Higher values indicate better results for the reported metric.

Source: Arena · CC BY 4.0. Redrawn from released data; reduced to selected models and rounded. Not affiliated with the source organization.

Explore related charts