List your company

Terminal Task Completion on Terminal-bench 2.0

64.7Pass@1Gemini 3 Pro
-1.423215.743432.9150.0766Feb 3, 2026Feb 26, 2026Mar 21, 2026Apr 13, 2026May 6, 2026May 29, 2026Jun 22, 2026

Evaluation results

MethodLinks
Gemini 3 ProOpenSource=false, Agent Framework=Best with Any Agent2026.0264.7————
GPT 5.2OpenSource=false, Agent Framework=Best with Any Agent2026.0262.9————
Claude Opus 4.5OpenSource=false, Agent Framework=Best with Any Agent2026.0257.8————
AORCHESTRAModel Setup=Gemini-3-Flash2026.0252.8657.1471.62——
GLM 5Model Size=744B2026.0652.4————
Claude Opus 4.5OpenSource=false, Agent Framework=OpenHands2026.0251.9————
MiniMax M2.7Model Size=230B2026.0645.1————
GPT-5OpenSource=false, Agent Framework=OpenHands2026.0243.8————
Kimi K2.5Model Size=1T2026.0643.2————
Claude Sonnet 4.5OpenSource=false, Agent Framework=Best with Any Agent2026.0242.8————
TMAX-27BModel Size=27B2026.0642.7————
Claude Sonnet 4.5OpenSource=false, Agent Framework=OpenHands2026.0242.6————
DeepSeek-v3.2Model Size=671B2026.0639.6————
Qwen 3.6Model Size=27B2026.0639.6————
Claude Opus 4.1OpenSource=false, Agent Framework=Best with Any Agent2026.0238————
Claude Opus 4.1OpenSource=false, Agent Framework=OpenHands2026.0236.9————
Minimax-M2.1OpenSource=true, Agent Framework=Best with Any Agent2026.0236.6————
AORCHESTRAModel Setup=Claude-4.5-haiku2026.0235.7145.7155.44——
Mini-SWEModel Setup=Gemini-3-Flash2026.0234.295049.49——
Claude CodeModel Setup=Claude-4.5-haiku2026.0234.2945.7129.65——
Claude CodeModel Setup=Gemini-3-Flash2026.0232.8648.5727.43——
Gemini 2.5 ProOpenSource=false, Agent Framework=Best with Any Agent2026.0232.6————
GPT-5-miniModel Size=?2026.0631.9————
OpenHandsModel Setup=Gemini-3-Flash2026.0231.4351.4348.49——
AORCHESTRAModel Setup=DeepSeek-V3.22026.0231.4342.8658.43——
LiberCoder-235B-A22BOpenSource=true, Agent Framework=OpenHands2026.0231————
LiberCoderModel Size=235B2026.0631————
Minimax-M2OpenSource=true, Agent Framework=Best with Any Agent2026.0230————
Mini-SWEModel Setup=DeepSeek-V3.22026.023048.5754.76——
Claude Haiku 4.5OpenSource=false, Agent Framework=Best with Any Agent2026.0229.8————
Claude Haiku 4.5Model Size=?2026.0629.8————
MTL (I)Model Backbone=DeepSeek V3.22026.0428.8————
ReActModel Setup=Gemini-3-Flash2026.0228.5747.1447.22——
ZeroShotModel Backbone=DeepSeek V3.22026.0428.5————
Kimi-K2-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.0227.8————
Nemotron-TerminalModel Size=32B2026.0627.4————
Grok 4OpenSource=false, Agent Framework=OpenHands2026.0227.2————
Grok 4OpenSource=false, Agent Framework=Best with Any Agent2026.0227.2————
Qwen3-Coder-480B-A35B-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.0227.2————
TMAX-9BModel Size=9B2026.0627.2————
Kimi-K2-InstructOpenSource=true, Agent Framework=OpenHands2026.0226.7————
Qwen3-Coder-480B-A35B-InstructOpenSource=true, Agent Framework=OpenHands2026.0225.4————
GLM-4.6OpenSource=true, Agent Framework=Best with Any Agent2026.0224.5————
MTL (W)Model Backbone=GPT-5-mini2026.0424.3————
Mini-SWEModel Setup=Claude-4.5-haiku2026.0224.2928.5736.3——
TerminalTraj-32BScaffold=Terminus-22026.0523.88———37.08
MTL (S)Model Backbone=GPT-5-mini2026.0422.8————
MTL (I)Model Backbone=Qwen3-Coder-480B-A35B-Instruct2026.0422.8————
TerminalTrajModel Size=32B2026.0622————
OpenHandsModel Setup=DeepSeek-V3.22026.0221.4335.7148.35——
Nemotron-Terminal-32BScaffold=Terminus-22026.0521.35———35.96
MTL (I)Model Backbone=GPT-5-mini2026.0421.3————
Qwen 3.5Model Size=9B2026.0621.1————
ZeroShotModel Backbone=Qwen3-Coder-480B-A35B-Instruct2026.0421————
ZeroShotModel Backbone=GPT-5-mini2026.0420.6————
Nemotron-TerminalModel Size=14B2026.0620.2————
ReActModel Setup=DeepSeek-V3.22026.022032.8638.23——
ReActModel Setup=Claude-4.5-haiku2026.022037.1443.62——
LiberCoder-32BOpenSource=true, Agent Framework=OpenHands2026.0219.5————
MTL (T)Model Backbone=GPT-5-mini2026.0419.5————
LiberCoderModel Size=32B2026.0619.5————
TermiGenModel Size=32B2026.0619.3————
TerminalTrajModel Size=14B2026.0619.1————
TMAX-4BModel Size=4B2026.0618.9————
GPT-OSS-120BOpenSource=true, Agent Framework=Best with Any Agent2026.0218.7————
GPT-OSSModel Size=120B2026.0618.7————
Qwen3-32B-Nex-N1Scaffold=OpenHands2026.0518.54———26.97
LiteCoder-Terminal-32b-sftScaffold=Terminus-22026.0518.54———30.34
Qwen3-235B-A22B-InstructOpenSource=true, Agent Framework=OpenHands2026.0218.1————
Qwen 3.5Model Size=4B2026.0616.6————
Gemini 2.5 ProOpenSource=false, Agent Framework=OpenHands2026.0216.4————
Claude Haiku 4.5OpenSource=false, Agent Framework=OpenHands2026.0213.9————
Nemotron-TerminalModel Size=8B2026.0613————
Qwen3-Coder-30B-A3B-InstructOpenSource=true, Agent Framework=OpenHands2026.0212.9————
Qwen3-Coder-30B-A3B-InstructOpenSource=true, Agent Framework=Best with Any Agent2026.0212.9————
OpenHandsModel Setup=Claude-4.5-haiku2026.0212.8525.7145.13——
Qwen3-30B-A3B-Nex-N1Scaffold=OpenHands2026.0512.36———23.6
LiteCoder-Terminal-30b-a3b-sftScaffold=Terminus-22026.0512.36———23.6
GPT-5-nanoModel Size=?2026.0611.5————
TerminalTrajModel Size=7B2026.0610.1————
EndlessTerminalsModel Size=8B, Training Mode=OT SFT + RL2026.066.7————
Qwen3-32BOpenSource=true, Agent Framework=OpenHands2026.025.7————
Qwen3-30B-A3B-InstructScaffold=Terminus-22026.055.34———11.24
OpenThinker-Agent-v1Model Size=8B2026.064.9————
LiteCoder-Terminal-4b-sftScaffold=Terminus-22026.054.78———10.11
Qwen2.5-Coder-32B-InstructScaffold=Terminus-22026.054.49———8.99
OpenThinker-Agent-v1Scaffold=Terminus-22026.054.49———10.1
TMAX-2BModel Size=2B2026.062.9————
Qwen 3.5Model Size=2B2026.062.3————
Qwen3-4B-InstructScaffold=Terminus-22026.051.12———3.37
BConfiguration=5 native flags, Backbone=gpt-5.4-nano2026.04———17—
BaselineConfiguration=no TURBOQUANT, Backbone=gpt-5.4-nano2026.04———15—
DConfiguration=all-on, 8 flags, Backbone=gpt-5.4-nano2026.04———12—
HARBORConfiguration=2 flags, Backbone=gpt-5.4-nano2026.04———17—
OracleConfiguration=union over configs, Backbone=gpt-5.4-nano2026.04———81—