Gemini 4 Neon 上线 Devin Desktop 和 Devin CLI,成本约为 Opus 5.5 的三分之一
Cognition 把 Gemini 4 Neon 接进了 Devin Desktop 和 CLI,实测 46.2% 分数几乎追平 Opus 5.5,价格只要三分之一。写代码跑 Agent 的团队可以直接换上试试,省钱不牺牲质量。
Cognition 宣布 Gemini 4 Neon 现已在 Devin Desktop 和 Devin CLI 中可用。在 FrontierCode 2.0 基准上,该模型得分 46.2%,与 Opus 5.5 的差距在 1 分以内,但每任务成本仅 $4.38,约为 Opus 5.5 的三分之一。
FrontierCode 2.0 是 Cognition 自家的基准,衡量标准是「这个 PR 是否应该被合并」,更贴近真实代码评审场景。Gemini 4 Neon 是 Gemini 系列在该基准上表现最强的模型,Cognition 称其性能与前沿模型相当。
对开发者来说,这意味着在 Devin 的工作流里可以低价获得接近 Opus 5.5 的代码质量,成本敏感的团队值得一试;Cognition 与 Google 的基准数据也来自一方评估,实际效果需结合自身任务验证。
Try Gemini 4 Neon in Devin Desktop and Devin CLI! We've found it to perform on par with frontier models on FrontierCode 2.0: Cognition @cognition We’ve evaluated Gemini 4 Neon on FrontierCode 2.0, our benchmark that measures whether you should merge the PR, and found that it scores within one point of Opus 5.5 at about a third of the cost. It is the strongest of the Gemini model series with a score of 46.2% and a cost of $4.38 per task. Gemini 4 Neon is now available to use in Devin Desktop and Devin CLI! 🔗 View Quoted Tweet📊 11 ⚡
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Gemini 4 Neon, Devin, Cognition, Opus 5.5.
Source attribution: curated from Windsurf (@windsurf_ai).