TAU2
Measured May 14, 2026Source
Score
0.75
Models15 1 
Grok 4
xAIxAIUnited States
Grok 4 is xAI's flagship large language model, designed for advanced reasoning, complex problem-solving, and code generation. It features multimodal capabilities, allowing it to process both text and images, and is optimized for fast response times.
ReasoningCodingFastMultimodal
Input / 1M tokens
$4.25Output / 1M tokens
$21.25Output tokens/s
43.01First-token seconds
17.08sSupported plans
1Benchmark history
Evaluations
Terminalbench Hard
Measured May 14, 2026Source
Score
0.38
Lcr
Measured May 14, 2026Source
Score
0.68
Ifbench
Measured May 14, 2026Source
Score
0.54
Aime 25
Measured May 14, 2026Source
Score
0.93
Aime
Measured May 14, 2026Source
Score
0.94
Math 500
Measured May 14, 2026Source
Score
0.99
Scicode
Measured May 14, 2026Source
Score
0.46
Livecodebench
Measured May 14, 2026Source
Score
0.82
Hle
Measured May 14, 2026Source
Score
0.24
Gpqa
Measured May 14, 2026Source
Score
0.88
Mmlu Pro
Measured May 14, 2026Source
Score
0.87
Artificial Analysis Math Index
Measured May 14, 2026Source
Score
92.7
Artificial Analysis Coding Index
Measured May 14, 2026Source
Score
40.5
Artificial Analysis Intelligence Index
Measured May 14, 2026Source
Score
41.5
Plan availability
Products and plans that support this model
ZenMux Builder
ZenMux Builder
ZenMux is a unified API gateway that provides access to over 100 AI models from providers like OpenAI, Anthropic, Google, and xAI through a single account and API. It offers subscription plans (Builder Plan) for personal development, vibe coding, and testing, as well as pay-as-you-go for production use. The platform supports coding agents like Claude Code and OpenClaw, and provides automatic model routing, multi-provider failover, and AI quality insurance compensation.
Discussion
Thinking... Make sure you are connected to GitHub server