Gemini 4 Carbon 在 Agent Arena 排名第 2，净提升 16.2%
Gemini 4 Carbon by @GoogleDeepmind is #2 in Agent Arena based on real-world tasks (+16.2% net improvement), and sits on the Pareto frontier for cost, time and o
Arena.ai 称 Gemini 4 Carbon 在基于真实任务的 Agent Arena 中排名第 2，净提升 16.2%，并在成本、时间和每任务输出 token 上处于帕累托前沿。相比排名第 7、提升 11% 的 Gemini 4 Argon，Carbon 高出 5 个名次和 5.2 个百分点，提升主要来自好评与投诉比 +13.2 个百分点、确认成功 +5.6 个百分点、可操控性 +4.8 个百分点。Arena 表示 Carbon 的可用性将由 Google DeepMind 公布。
推荐理由
精选AI 评分 70
Arena 的实测数据把 Carbon 相对 Argon 的提升拆到可操控性、任务完成率和用户情绪三个信号上，便于判断差距来源。
主题
查看原帖激动人心的消息：来自 @GoogleDeepMind 的 Gemini 4 Carbon (High) 刚刚在 Arena 上完成重磅首秀，在 Text Arena 排名第 1，在 Agent Arena 排名第 2，在 Code Arena: WebDev 排名第 4。Gemini 4 Carbon 重返榜首，带来前沿智能，并在三大竞技场重塑帕累托前沿。 在 Agent Arena 中，Gemini 4 Carbon 排名第 2，在我们全球社区超过 2.19 万次真实世界智能体会话中实现 +16.2% 的净提升，超越顶尖模型 GPT-6 Astra 和 Fable-5.1。Gemini 4 Carbon (High) 每任务成本中位数为 1.80 美元，比排名第 1 的 Claude Opus 5.5 (Max) 低 72%。 其关键信号亮点： - 可操控性第 1（+18.6%）：当你提出异议时，模型能否及时纠正方向？ - 好评与投诉比第 2（+37.9%）：用户反应中的隐含情绪 - 确认成功第 3（+18.7%）：用户明确表示“是的，这有效” 在 Code Arena: WebDev 中，Carbon 获得 1768 分，排名第 4： - 仅比排名第 3 的 Claude Sonnet 5.5 (xHigh) 低 2 分，后者为 1770 分 - 比排名第 11 的 Gemini 4 Argon (High) 高 89 分，后者为 1679 分 按类别划分，它在游戏与模拟中排名第 3，在品牌与营销中排名第 5。 在 Text Arena 中，Gemini 4 Carbon (High) 以 1528 分成为新的第 1，领先于以 1520 分排名第 2 的 Gemini 4 Argon (High)。 祝贺 @GoogleDeepMind 团队达成这一重大里程碑！
查看原文
Exciting news: Gemini 4 Carbon (High) from @GoogleDeepMind just made a huge debut on Arena, landing #1 in Text Arena, #2 in Agent Arena, and #4 in Code Arena: WebDev. Back at the top, Gemini 4 Carbon delivers frontier intelligence and reshapes the Pareto frontier across three arenas. In Agent Arena, Gemini 4 Carbon ranks #2 with a +16.2% net improvement across +21.9K real-world agentic sessions from our global community, surpassing top models GPT-6 Astra and Fable-5.1. At $1.80 median cost per task, Gemini 4 Carbon (High) costs 72% less than the #1 ranked Claude Opus 5.5 (Max). Its standouts by key signals: - #1 Steerability (+18.6%): can the model course-correct when you push back? - #2 Praise vs Complaint (+37.9%): implicit sentiment in users' reactions - #3 Confirmed Success (+18.7%): explicit “yes, that worked” from users In Code Arena: WebDev, Carbon scored 1768 pts landing #4: - Just 2 pts below #3 Claude Sonnet 5.5 (xHigh), at 1770 pts - 89 pts above #11 Gemini 4 Argon (High), at 1679 pts By category, it ranks #3 in Gaming and Simulations, and #5 in Brand & Marketing. In Text Arena, Gemini 4 Carbon (High) is the new #1 with 1528 pts, now ahead of Gemini 4 Argon (High) at #2 with 1520 pts. Congrats to the @GoogleDeepMind team on this huge milestone!
正文
Gemini 4 Carbon by @GoogleDeepmind is #2 in Agent Arena based on real-world tasks (+16.2% net improvement), and sits on the Pareto frontier for cost, time and output tokens per task!
This release is 5 rankings and +5.2 percentage pts higher performing than Gemini 4 Argon (#7 with +11%).
Agent Arena signals show that Carbon’s gains over Argon are coming from:
1. Receives markedly more praise than the previous release (Praise vs Complaint: +13.2 pp)
2. Completes tasks more often (Confirmed Success: +5.6 pp)
3. Handles your feedback better (Steerability: +4.8 pp)
Stay tuned for Carbon availability from @googledeepmind, and dive into the Agent Arena at the link in thread.