GPT
OpenAI
United States
OpenAI's flagship reasoning model
Official site ↗PROJECT HORIZON · LLM JURY
The LLM Jury
Twenty AI models from three continents, asked the same question on the same day — and giving sometimes very different answers.
The panel
19 The LLM Jury
Claude
Anthropic
United States
Anthropic's safety-first frontier model
Official site ↗
Gemini
Google DeepMind
United States
Google's multimodal flagship
Official site ↗
Grok
xAI
United States
xAI's assistant by Elon Musk's lab
Official site ↗
Pe
Perplexity
Perplexity AI
United States
Search-augmented question answering
Official site ↗
Llama
Meta
United States
Meta's open weights family
Official site ↗
Co
Command R
Cohere
Canada
Cohere's enterprise-tuned model
Official site ↗
Ph
Phi
Microsoft
United States
Microsoft's small efficient models
Official site ↗
Mistral
Mistral AI
France
The French open frontier challenger
Official site ↗
Le
Le Chat
Mistral AI
France
Mistral's conversational assistant
Official site ↗
Qwen
Alibaba
China
Alibaba's Tongyi Qianwen family
Official site ↗
DeepSeek
DeepSeek AI
China
High-performance reasoning models
Official site ↗
Kimi
Moonshot AI
China
Moonshot's long-context champion
Official site ↗
ERNIE
Baidu
China
Baidu's Wenxin Yiyan flagship
Official site ↗
GL
GLM
Zhipu AI
China
Tsinghua's bilingual specialist
Official site ↗
Yi
Yi
01.AI
China
Kai-Fu Lee's open weights
Official site ↗
Hu
Hunyuan
Tencent
China
Tencent's multimodal model
Official site ↗
Doubao
ByteDance
China
ByteDance's flagship LLM
Official site ↗
MiniMax
MiniMax
China
High-context Chinese frontier model
Official site ↗