Skip to content
AIMODELRACE
Build software

Best agent for devops & infrastructure

CI/CD, containers, cloud configuration, monitoring and incident response.

Leader

Photo finish

Claude Sonnet 5.5

Index

93

Lead

+0.6

% ahead

0.6%

over Claude Opus 5.5. Too close to call: choose on price, speed or the tools you already use.

Full ranking

Task index, 0 to 100. Bars are scaled to the field, so small gaps look small.

  1. 1
    Claude Sonnet 5.5
    93

    Leads terminal and shell work.

  2. 2
    Claude Opus 5.5
    −0.692.4

    Better for incident diagnosis.

  3. 3
    GPT-6 Astra
    −5.088

    Strong across cloud providers.

  4. 4
    GLM-5.3
    −9.084

    Best open option for terminal agents.

  5. 5
    Grok 4.7
    −11.082

    Capable and cheap.

Key takeaways
  • 1Sonnet 5.5 leads terminal-heavy work in our races.
  • 2Never give agents unrestricted production credentials.
  • 3Infrastructure-as-code makes agent changes reviewable.

Recommended setup

A coding agent with read-only production access, write access only to IaC repos, and changes deployed through your normal pipeline.

Prompts that work

CI failure fix

Our CI fails at {{step}}. Log: {{log}}. Diagnose root cause, propose the minimal fix, and give exact commands to reproduce and verify locally. Do not change unrelated config.

More in build software

Frontend & UI

Claude Opus 5.5 (Claude Code)+2.0

Backend & APIs

Claude Opus 5.5+1.0

Distributed systems

Claude Fable 5.1+4.0

Website building

Claude Code + Opus 5.5+2.0

App building

Claude Code + Opus 5.5+3.0

Data & analytics

GPT-6 Astra+0.7

Code review & security

Claude Opus 5.5+3.0