All benchmarks
xAI · USA · Released 2025-07
Grok 4
xAI's frontier reasoning model with real-time X data grounding and a heavy multi-agent variant (Grok 4 Heavy).
Access
closed
Context
256,000 tokens
Modalities
text, vision
Languages
80+
Benchmark scores
MMLU
87.0%
Massive Multitask Language Understanding
GPQA
87.5%
Graduate-level science QA (Diamond)
HumanEval
92.0%
Python code completion
SWE-bench
N/A
Real GitHub issue resolution (Verified)
LiveCodeBench
N/A
Contamination-free coding
AIME
95.0%
American Invitational Mathematics Exam
MMMU
78.0%
Multimodal college-level reasoning
MATH
93.0%
Competition-level math word problems
Scores as reported by the vendor or leading public leaderboards. "N/A" means the score has not been publicly disclosed for this metric.
Explain like I'm 5
Very good at hard exams and knows what's happening on X right now.
Key features
- Real-time X grounding
- Multi-agent Heavy mode
- Tool use
Strengths
- Top-tier on Humanity's Last Exam
- Fresh real-time knowledge
Limitations
- Limited multimodal breadth
- Closed source
Best for
Real-time researchReasoning-heavy tasks
Related models
OpenAI
GPT-5
OpenAI's flagship reasoning + multimodal model with a unified router that switches between fast and deep-thinking modes.
Anthropic
Claude Sonnet 4.5
Anthropic's best coding + agentic model, purpose-built for long-horizon computer-use and software engineering tasks.
Google DeepMind
Gemini 2.5 Pro
Google's most capable model with native 1M-token context and full multimodal I/O across text, images, audio and video.
Meta AI
Llama 4 Maverick
Meta's flagship MoE model — 400B total / 17B active — with native multimodal input and 1M-token context.