TAU2
Measured Jul 10, 2026Source
Score
0.93
Models8 1 
Grok 4.20 0309 v2 (Reasoning)
xAIxAIUnited States
Grok 4.20 is a reasoning-focused model from xAI, optimized for complex problem-solving and logical inference. It delivers fast response times while maintaining strong performance on analytical tasks, offering a cost-effective solution for advanced reasoning applications.
ReasoningFastCheapLong context
Input / 1M tokens
$2.00Output / 1M tokens
$6.00Output tokens/s
164.43First-token seconds
16.84sSupported plans
2Benchmark history
Evaluations
Terminalbench Hard
Measured Jul 10, 2026Source
Score
0.38
Lcr
Measured Jul 10, 2026Source
Score
0.58
Ifbench
Measured Jul 10, 2026Source
Score
0.81
Scicode
Measured Jul 10, 2026Source
Score
0.46
Hle
Measured Jul 10, 2026Source
Score
0.32
Gpqa
Measured Jul 10, 2026Source
Score
0.91
Artificial Analysis Intelligence Index
Measured Jul 10, 2026Source
Score
37
Plan availability
Products and plans that support this model
NanoGPT
NanoGPT
NanoGPT is a privacy-focused AI platform that provides unified access to multiple text, image, video, and audio AI models through a chat interface and API, with features like auto model selection and pay-as-you-go pricing.
User ratings
Loading ratings...
Discussion
Thinking... Make sure you are connected to GitHub server