Artificial Analysis 发布 Coding Agent Index v2.1，Claude 双模型并列第一
给选模型的开发者一个新榜单：Anthropic 两款 Claude 并列 58 分第一，但每任务成本 $44.22，是 GPT-6.1 Sol（$1.51）的近 30 倍。预算敏感的团队可以重点看 Codex 系的性价比表现。
Artificial Analysis 发布 Coding Agent Index v2.1，将基准 FrontierCode 从 1.1 升级到 2.0、Terminal-Bench 从 4.0 升级到 4.1。FrontierCode-AA 2.0 包含 132 个真实编码任务，以仓库维护者是否会合并该改动作为评分标准；Terminal-Bench 4.1 保留 66 个任务，更新了验证器以提升可靠性和抗 reward hacking 能力。
榜单上 Claude Fable 5.2 与 Claude Opus 5.5 以 58 分并列第一：前者在 Claude Code 的 xhigh 和 high 档均得 58，后者在 Claude Code（max 档）以及搭配 GPT-6 Luna High 的 Devin Fusion CLI 多模型环境中均得 58。Gemini 4 Carbon（high，Antigravity CLI）以 55 分紧随其后。
由于 FrontierCode-AA 2.0 更严格，v2.1 的分数整体下降约 5 分，唯一基本未跌的是 Codex with GPT-6.1 Terra（max）。成本方面，48 分以上配置的单任务花费相差约 30 倍：GPT-6.1 Sol（max，52 分）为 $1.51，而 Claude Fable 5.2（max，56 分）在 Claude Code 中高达 $44.22。
Announcing Artificial Analysis Coding Agent Index v2.1, upgrading FrontierCode and Terminal-Bench. Claude Fable 5.2 and Claude Opus 5.5 share the lead at 58
The Coding Agent Index measures models in agent harnesses; v2.1 of the Index upgrades FrontierCode from 1.1 to 2.0 and Terminal-Bench from 4.0 to 4.1.
Upgraded:
➤ FrontierCode-AA 2.0 (replacing FrontierCode v1.1): our implementation of Cognition's FrontierCode 2.0, comprising 132 real-world coding tasks graded on whether a repository maintainer would merge the agent's change
➤ Terminal-Bench 4.1 (replacing Terminal-Bench 4.0): the same 66 tasks, scored with updated verifiers that improve reliability and resistance to reward hacking
Staying in the Index: SWE-Atlas-QnA-AA
Key results:
➤ Claude Fable 5.2 and Claude Opus 5.5 share the lead at 58: Claude Fable 5.2 scores 58 at both xhigh and high in Claude Code. Claude Opus 5.5 (max) scores 58 in Claude Code and, paired with GPT-6 Luna High, in Cognition's multi-model harness Devin Fusion CLI. Claude Fable 5.2 (max) scores 56 as more of its attempts hit task time limits. Claude Sonnet 5.5 (max, 56) in Claude Code and Gemini 4 Carbon (high, 55) in Antigravity CLI follow
➤ Index scores are lower in v2.1, driven by the upgrade to FrontierCode-AA 2.0: models score lower by 5 points on average. Codex with GPT-6.1 Terra (max) is the only configuration which didn’t fall between versions, sitting within 1 point of its prior score
➤ Cost per task varies ~30x among configurations scoring 48+: from $1.51 for Codex with GPT-6.1 Sol (max, 52) to $44.22 for Claude Code with Claude Fable 5.2 (max, 56)
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 US-China foundation model benchmarks and commercialization comparison and AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Claude, Codex, Gemini, Terminal-Bench.
Source attribution: curated from Artificial Analysis (@ArtificialAnlys).
