Model Breakdowns

Version-by-version analysis of frontier AI models — context windows, benchmarks, and highlights.

Quick Comparison

ModelVendorMax ContextVersionsMMLU (top)HumanEval (top)
GLM-5.1Zhipu AI / Zai-Org128K tokens388.492.1
Claude 4Anthropic200K tokens389.292.5
GPT-5OpenAI256K tokens390.194.2
Gemini 2.5 ProGoogle DeepMind2M tokens387.890.4
DeepSeek V4DeepSeek128K tokens387.589.8