EN中
← All visual benchmarksArena / 2026-10-22
Arena · INSTRUCTION

Instruction following

How well responses follow instructions in human comparisons.

Hover or tap a bar to inspect. Select up to 3 models to highlight and compare.
Model scores10 models
Instruction following. Zoomed horizontal axis, from 1485.7 to 1545.7ZOOMED X-AXIS · 1485.7 – 1545.7 · NOT FROM ZERO1485.71500.71515.71530.71545.7Claude Opus 5.5 (High)Claude Opus 5.5 (High) · 1538.8 Arena Rating · zoomed horizontal axis1538.8Gemini 4 Argon (High)Gemini 4 Argon (High) · 1527.6 Arena Rating · zoomed horizontal axis1527.6Claude Opus 4.6 (High)Claude Opus 4.6 (High) · 1521 Arena Rating · zoomed horizontal axis1521Claude Fable 5.1 (Max)Claude Fable 5.1 (Max) · 1518.9 Arena Rating · zoomed horizontal axis1518.9Claude Opus 5 (Max)Claude Opus 5 (Max) · 1518.4 Arena Rating · zoomed horizontal axis1518.4Claude Opus 5 (High)Claude Opus 5 (High) · 1514.2 Arena Rating · zoomed horizontal axis1514.2Claude Opus 4.6Claude Opus 4.6 · 1509.9 Arena Rating · zoomed horizontal axis1509.9Claude Fable 5 (High)Claude Fable 5 (High) · 1501.6 Arena Rating · zoomed horizontal axis1501.6Claude Opus 4.7 (High)Claude Opus 4.7 (High) · 1497.7 Arena Rating · zoomed horizontal axis1497.7Muse Wonder (Max)Muse Wonder (Max) · 1492.6 Arena Rating · zoomed horizontal axis1492.6
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 all benchmark charts