영국의 AI 보안 연구소는 표준 벤치마크가 AI 에이전트가 실제로 수행할 수 있는 작업을 체계적으로 과소평가한다는 사실을 발견했습니다. | HotON.ai

물가

MINIMAX-M2.1$0.951000K

GEMINI-3.1-FLA$1.501049K

GPT-4.1-MINI$1.601048K

GEMINI-2.0-FLA$0.401049K

DEEPSEEK$0.281049K

GROK-4-1-FAST-$0.502000K

MINIMAX-M2$1.001000K

LLAMA-4-MAVERI$0.601049K

GEMINI-2.5-FLA$0.401049K

MINIMAX-M3$1.201000K

QWEN-PLUS$0.781000K

GEMINI-2.5-FLA$0.401049K

QWEN3.5-FLASH-$0.261000K

GEMINI-2.0-FLA$0.301049K

GROK-4.1-FAST$0.502000K

QWEN3-CODER-FL$0.981000K

GPT-4.1-NANO$0.401048K

QWEN-PLUS-2025$0.781000K

GROK-4-FAST$0.502000K

QWEN-PLUS-2025$0.781000K

DEEPSEEK-V4-FL$0.281049K

MINIMAX-M2.7$1.201000K

GROK-4-1-FAST-$0.502000K

MINIMAX-01$1.101000K

GROK-4-FAST-RE$0.502000K

GPT-4.1-MINI-2$1.601048K

MINIMAX-M2.5$0.991000K

GROK-4-FAST-NO$0.502000K

MINIMAX-M2.1$0.951000K

GEMINI-3.1-FLA$1.501049K

GPT-4.1-MINI$1.601048K

GEMINI-2.0-FLA$0.401049K

DEEPSEEK$0.281049K

GROK-4-1-FAST-$0.502000K

MINIMAX-M2$1.001000K

LLAMA-4-MAVERI$0.601049K

GEMINI-2.5-FLA$0.401049K

MINIMAX-M3$1.201000K

QWEN-PLUS$0.781000K

GEMINI-2.5-FLA$0.401049K

QWEN3.5-FLASH-$0.261000K

GEMINI-2.0-FLA$0.301049K

GROK-4.1-FAST$0.502000K

QWEN3-CODER-FL$0.981000K

GPT-4.1-NANO$0.401048K

QWEN-PLUS-2025$0.781000K

GROK-4-FAST$0.502000K

QWEN-PLUS-2025$0.781000K

DEEPSEEK-V4-FL$0.281049K

MINIMAX-M2.7$1.201000K

GROK-4-1-FAST-$0.502000K

MINIMAX-01$1.101000K

GROK-4-FAST-RE$0.502000K

GPT-4.1-MINI-2$1.601048K

MINIMAX-M2.5$0.991000K

GROK-4-FAST-NO$0.502000K