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

Chinese conversations

Chinese dialogue preference, not translation quality.

Hover or tap a bar to inspect. Select up to 3 models to highlight and compare.
Model scores10 models
Chinese conversations. Zoomed horizontal axis, from 1535.0 to 1608.7ZOOMED X-AXIS · 1535.0 – 1608.7 · NOT FROM ZERO1535.01553.41571.81590.31608.7Gemini 4 Argon (High)Gemini 4 Argon (High) · 1600.2 Arena Rating · zoomed horizontal axis1600.2Claude Fable 5.1 (Max)Claude Fable 5.1 (Max) · 1583.6 Arena Rating · zoomed horizontal axis1583.6Claude Opus 5.5 (High)Claude Opus 5.5 (High) · 1581 Arena Rating · zoomed horizontal axis1581Claude Opus 5 (Max)Claude Opus 5 (Max) · 1573.1 Arena Rating · zoomed horizontal axis1573.1Claude Opus 5 (High)Claude Opus 5 (High) · 1567.4 Arena Rating · zoomed horizontal axis1567.4Gemini 3.7 Flash (High)Gemini 3.7 Flash (High) · 1553.1 Arena Rating · zoomed horizontal axis1553.1Gemini 3.8 Flash (High)Gemini 3.8 Flash (High) · 1552.5 Arena Rating · zoomed horizontal axis1552.5Claude Opus 4.6 (High)Claude Opus 4.6 (High) · 1551 Arena Rating · zoomed horizontal axis1551Claude Opus 4.6Claude Opus 4.6 · 1546.5 Arena Rating · zoomed horizontal axis1546.5Claude Fable 5 (High)Claude Fable 5 (High) · 1543.5 Arena Rating · zoomed horizontal axis1543.5
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