specific method · filed under software implementation
Claude Code
A coding-agent scaffold used to run evaluations and described as a backend-compatible scaffold.
- sources
- 3
- models
- 4
- labs adopt it
- 3
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
optional 1used 2
Documented in
Further reading
Picked by hand, not extracted: where to read more, not evidence for anything on this page.
- Claude Code overview docs code.claude.comofficial documentation for the agentic coding tool
Evidence
3 spans quoted from the sources, strongest treatment first.
SWE-Bench Verified: Tested on internal infrastructure using Claude Code as the scaffolding.
usedsoftware implementationin MiniMax-M3Anthropic
We evaluate Terminal-Bench 2.1 with Claude Code 2.1.167
usedevaluation onlyin GLM-5.2Z.ai
Xiaomi explicitly positions it as a drop-in backend for scaffolds including Claude Code, OpenCode, and Kilo
optionalsoftware implementationin MiMo-V2.5-ProXiaomi
Filed alongside
Other methods under software implementation :: agent product.