ARAgentRanks
~ / stacks / claude-code-sonnet-4-6

Claude Code + Sonnet 4.6

A CLI agent plus hosted coding model stack for deep terminal workflows and complex multi-step refactors.

ARscore 68.51M context$2/$8mixed stack
AgentClaude Code
ModelSonnet 4.6
Agent score86
Model SWE-bench79.6%
Combined score68.5
Best fitdeep terminal workflows and complex multi-step refactors

Related Claude Code stacks

Same agent, different model choices.

Claude Code + Opus Mythos

ARscore 80.8, 1M context, TBD

view stack

Claude Code + Opus 4.8

ARscore 76.2, 1M context, $5/$25

view stack

Claude Code + Opus 4.7

ARscore 75.3, 1M context, $5/$30

view stack

Related Sonnet 4.6 stacks

Same model, different agent wrappers.

Devin + Sonnet 4.6

ARscore 48.6, Cloud workflow

view stack

Continue + Sonnet 4.6

ARscore 47, Extension workflow

view stack

Windsurf + Sonnet 4.6

ARscore 47, IDE workflow

view stack

Oh-My-Codex + Sonnet 4.6

ARscore 37.4, CLI workflow

view stack