List your company

Agentic Coding on TerminalBench 2

81.8Pass RateForgeCode
11.18429.51747.8566.183Mar 30, 2026

Evaluation results

MethodLinks
ForgeCodeBase Model=Claude Opus 4.6, Auto=false2026.0381.8
Meta-HarnessBase Model=Claude Opus 4.6, Auto=true2026.0376.4
CapyBase Model=Claude Opus 4.6, Auto=false2026.0375.3
Terminus-KIRABase Model=Claude Opus 4.6, Auto=false2026.0374.7
MAYA-V2Base Model=Claude Opus 4.6, Auto=false2026.0372.1
TongAgentsBase Model=Claude Opus 4.6, Auto=false2026.0371.9
DroidBase Model=Claude Opus 4.6, Auto=false2026.0369.9
MuxBase Model=Claude Opus 4.6, Auto=false2026.0366.5
Terminus 2Base Model=Claude Opus 4.6, Auto=false2026.0362.9
Claude CodeBase Model=Claude Opus 4.6, Auto=false2026.0358
Meta-HarnessBase Model=Claude Haiku 4.5, Auto=true2026.0337.6
GooseBase Model=Claude Haiku 4.5, Auto=false2026.0335.5
Terminus-KIRABase Model=Claude Haiku 4.5, Auto=false2026.0333.7
Mini-SWE-AgentBase Model=Claude Haiku 4.5, Auto=false2026.0329.8
Terminus 2Base Model=Claude Haiku 4.5, Auto=false2026.0328.3
Claude CodeBase Model=Claude Haiku 4.5, Auto=false2026.0327.5
OpenHandsBase Model=Claude Haiku 4.5, Auto=false2026.0313.9