agent profile
Agent rank#1
Score91.3
Best on Aider: gpt-oss-120bopen · 100
Best models· Aider
14 models · CodingAgentBench
#Model
196%100276%89372%86468%84568%84668%84768%84852%73936%651020%591116%561212%54134%41140%34
gpt-oss-120b
OpenAIopen
llama-4-maverick-17b-128e-instruct
Metaopen
step-3.5-flash
StepFunopen
nemotron-3-super-120b-a12b
NVIDIAopen
GLM-5.1
Z.aiopen
MiniMax M2.7
MiniMaxopen
nemotron-3-ultra-550b-a55b
NVIDIAopen
llama-3.3-nemotron-super-49b-v1
NVIDIAopen
mistral-small-4-119b-2603
Mistral AIopen
llama-3.3-70b-instruct
Metaopen
qwen3.5-397b-a17b
Alibabaopen
qwen3.5-122b-a10b
Alibabaopen
qwen3-next-80b-a3b-instruct
Alibabaopen
MiniMax M3
MiniMaxopen