Sonnet 5.5 tops Coding Agent Index; xhigh delivers 93% of the score for 23% of the cost
ArtificialAnlys · x · 2026-10-14
Artificial Analysis updated its Coding Agent Index: Sonnet 5.5 in Claude Code tops the leaderboard at 68 (max effort), while the xhigh setting scores 63 at a fraction of the cost.
- Cost: $14.19/task at max vs $3.33 at xhigh — 77% savings for only 5.5 fewer index points
- Max generates 4.8× more output tokens and takes 3.4× more turns (206 turns, 1.5h per task)
- Max improves all three component benchmarks (DeepSWE v1.1, Terminal-Bench 4.0, SWE-Atlas-QnA), with the biggest gain on Terminal-Bench 4.0: 58.1% → 66.2%
- The board also features Muse Wonder (Meta), Antigravity CLI with Gemini 4 Argon (Google), Codex GPT-6.1 Sol (OpenAI), plus Kimi K3, GLM-5.3 and Qwen3.8 Max
More from coding & agent
- 10 Open-Source GitHub Repos That Give Your AI Agent Memory, from Mem0 (67K Stars) to Engram — victor_explore · 2026-10-14
- Free Bots: A Persistent 3D City Where AI Agents Actually Live, Work, and Build Houses — Daniel_Farinax · 2026-10-14
- Caught in the logs: Codex quietly delegated work to DeepSeek Flash — dotey · 2026-10-14
- Gemini 4 Argon lands on Devin at 25% off, mocked for a self-own benchmark chart — ChrisUniverse · 2026-10-14
- Indie dev builds code-native design canvas that works with any agent, no Figma MCP gate — talkaboutdesign · 2026-10-14
- 10 GitHub repos that give your AI agent memory: Mem0, Hindsight, Graphiti and more — victor_explore · 2026-10-14