List your company

Terminal Agent Tasks on TerminalBench 2.0

57.8Success ScoreClaude-Opus-4.5
0.28815.21930.1545.081Jun 22, 2026

Evaluation results

MethodLinks
Claude-Opus-4.5Size=-, Scaffold=Terminus 22026.0657.8———
Gemini 3 Pro PreviewSize=-, Scaffold=Terminus 22026.0656.9———
GPT-5.2Size=-, Scaffold=Terminus 22026.0654———
GLM-4.7Size=358B, Scaffold=Terminus 22026.0641———
CLI-Universe-32BSize=32B, Scaffold=Terminus 22026.0633.4———
LiberCoder-235BSize=235B, Scaffold=OpenHands2026.0631———
SkillSynth-32BSize=32B, Scaffold=Terminus 22026.0629.6———
Minimax-M2.1Size=229B, Scaffold=Terminus 22026.0629.2———
Kimi-K2-InstructSize=1T, Scaffold=Terminus 22026.0627.8———
Nemotron-Terminal-32BSize=32B, Scaffold=Terminus 22026.0627.4———
Qwen3-Coder-480BSize=480B, Scaffold=Terminus 22026.0623.9———
CLI-Universe-14BSize=14B, Scaffold=Terminus 22026.0623———
TerminalTraj-32BSize=32B, Scaffold=Terminus 22026.0622———
Nemotron-Terminal-14BSize=14B, Scaffold=Terminus 22026.0620.2———
SkillSynth-14BSize=14B, Scaffold=Terminus 22026.0619.9———
LiberCoder-32BSize=32B, Scaffold=OpenHands2026.0619.5———
TerminalTraj-14BSize=14B, Scaffold=Terminus 22026.0619.1———
Qwen3-Coder-30B-A3BSize=30B, Scaffold=Terminus 22026.0614.6———
SkillSynth-8BSize=8B, Scaffold=Terminus 22026.0613.5———
Nemotron-Terminal-8BSize=8B, Scaffold=Terminus 22026.0613———
CLI-Universe-8BSize=8B, Scaffold=Terminus 22026.0610.9———
TerminalTraj-7BSize=7B, Scaffold=Terminus 22026.0610.1———
Qwen3-14BSize=14B, Scaffold=Terminus 22026.064———
Qwen3-32BSize=32B, Scaffold=Terminus 22026.063.4———
Qwen3-8BSize=8B, Scaffold=Terminus 22026.062.5———
OT-SFTSetup=SFT2026.05—5.6210.4512.36
OT-SFTSetup=GRPO2026.05—7.6414.3817.98
OT-SFTSetup=ECHO2026.05—7.8713.8217.98
Qwen3-14BSetup=Base2026.05—4.278.9912.36
Qwen3-14BSetup=GRPO2026.05—5.1710.6713.48
Qwen3-14BSetup=ECHO2026.05—10.7916.5219.1
Qwen3-8BSetup=Base2026.05—1.573.714.49
Qwen3-8BSetup=GRPO2026.05—2.76.748.99
Qwen3-8BSetup=ECHO2026.05—5.1710.4513.48