Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Gemini 4 Carbon Ties for Lead on Arena Weekly Leaderboard

AI summary

IT之家 reports that Arena’s Week 46 anonymous blind-test leaderboard for November 9–15, 2026, places Gemini 4 Carbon (High) first with 1526 points and Gemini 4 Argon (High) second with 1522. Because the four-point gap is within the reported error range, IT之家 treats them as tied for the lead. Anthropic models occupy five of the top eight overall positions, while Dreamina Seedance 2.0, Dreamina Seedance 2.5 and Wan 3.0 rank sixth through eighth on the AI video leaderboard. The rankings are based on anonymous votes and reflect user preferences rather than absolute model capability; separate leaderboards should not be compared directly.

In-depth insight

The weekly results provide a current snapshot of user preferences across general, coding, agent, image and video models, with reported strong placements for Anthropic, Google and Chinese models.

IT之家 11 月 16 日消息,Arena(arena.ai)平台发布了 2026 年第 46 周(2026 年 11 月 9 日 ~11 月 15 日)大模型盲测投票榜单。

本周最大看点来自谷歌:新模型 Gemini 4 Carbon (High) 首秀便以 1526 分登顶综合榜,同家族的 Gemini 4 Argon (High) 以 1522 分紧随其后,两者仅差 4 分,在误差范围内视为并列第一。

国产模型方面,前端开发榜上月之暗面 kimi k3.1 pro (Max) 首次杀入第 12 名,成为榜单中排名最高的国产模型;AI 视频榜上,字节跳动 dreamina seedance 2.0 / 2.5 分别占据第 6、7 名,阿里 wan 3.0 位列第 8 名,国产阵营在视频生成赛道已稳定占据第一梯队。

整体来看,本周榜单呈现“谷歌新王登顶、Anthropic 固守智能体与前端、国产模型在多模态赛道持续渗透”的格局。

IT之家注:本榜单基于 Arena(arena.ai)平台匿名盲测投票,通过 ELO(智能体榜为百分比信号)计算排名,反映用户主观偏好而非绝对能力测试;不同分榜相互独立,不宜跨榜比较,分差在误差范围内均视为并列,新模型也需积累一定投票量后才会正式入榜。

#Arena#Gemini 4 Carbon#Gemini 4 Argon#Claude Opus 5.5#Claude Fable 5.2#GPT 6 Astra#Dreamina Seedance 2.0#Dreamina Seedance 2.5#Wan 3.0#Kimi K3.1 Pro#Qwen 3.8 Max#GLM 5.3

Original · 9 min readRead the original
Group chat

No comments yet. Start the conversation.