agent profile
SolidCLIin switcher
Copilot CLI
GitHub·14 models tested·CodingAgentBench
Agent rank#8
Score79.9
Best on Copilot CLI: step-3.5-flashopen · 96
Best models· Copilot CLI
14 models · CodingAgentBench
#Model
184%96276%89356%76456%76548%71632%64720%59812%54912%54104%41114%41120%34130%34140%34
step-3.5-flash
StepFunopen
gpt-oss-120b
OpenAIopen
mistral-small-4-119b-2603
Mistral AIopen
qwen3-next-80b-a3b-instruct
Alibabaopen
nemotron-3-super-120b-a12b
NVIDIAopen
qwen3.5-397b-a17b
Alibabaopen
GLM-5.1
Z.aiopen
llama-3.3-70b-instruct
Metaopen
qwen3.5-122b-a10b
Alibabaopen
llama-3.3-nemotron-super-49b-v1
NVIDIAopen
MiniMax M2.7
MiniMaxopen
llama-4-maverick-17b-128e-instruct
Metaopen
MiniMax M3
MiniMaxopen
nemotron-3-ultra-550b-a55b
NVIDIAopen