전문가 수준의 프롬프트에서의 사용자 선호도를 집계한 Arena Elo 점수입니다.
Anthropic
Claude Opus 4.7
Claude Opus 4.6
OpenAI
GPT-5.4
GPT-5.4 Pro
Google
Gemini 3.1 Pro
Xiaomi
MiMo-V2-Pro
Claude Opus 4.5
Claude Sonnet 4.6
Claude Sonnet 4.5
Gemini 3 Flash
Moonshot AI
Kimi K2.6
GPT-5.4 Mini
Kimi K2.5
Z.ai
GLM-5.1
Meta
Muse Spark
Alibaba
Qwen3.6 Plus
Claude Opus 4.1
Gemma 4 31B
GLM-5
DeepSeek
DeepSeek V4 Pro
Grok
Grok 4.20
Grok 4.20 (Reasoning)
Qwen3.5 397B A17B
Gemini 2.5 Pro
GPT-5
DeepSeek V4 Flash
Meituan
Longcat Flash Chat
Baidu
ERNIE 5.0 Thinking
GPT-5.4 Nano
Gemini 3.1 Flash Lite
Claude Haiku 4.5
DeepSeek V3.2
Claude Opus 4
MiniMax
MiniMax M2.5
Grok 4.1 Fast
Grok 4.1 Fast (Reasoning)
Claude Sonnet 4
MiniMax M2.7
Gemini 2.5 Flash
Arcee AI
Trinity Large Thinking
GPT-5 Mini
NVIDIA
Nemotron 3 Super
Gemini 2.5 Flash Lite
Amazon
Nova 2 Lite
GPT OSS 120B
GPT-5 Nano
Llama 4 Maverick
Llama 4 Scout
GPT-4.1