Grok 4
grok-4
ProprietaryMultimodal
About
Enhanced reasoning with long-form logic; multimodal support; live browsing and long-term memory.
Grok 4 has a 256K-token context window.
Grok 4 input tokens at $3/1M, output at $15/1M.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode ExecutionPrompt CachingBatch APIAudioFine-tuning
Providers(3)
Compare all →| Provider | Input (per 1M) | Output (per 1M) | Type | |
|---|---|---|---|---|
| Microsoft Foundry | — | — | ServerlessProvisioned | |
| OpenRouter | $3 | $15 | Serverless | |
| Replicate API | $7.2 | $36 | Serverless |
Benchmark Scores(4)
| Benchmark | Score | Version | Source |
|---|---|---|---|
| Aider Polyglot | 79.6 | 2026-04 (high) | https://aider.chat/docs/leaderboards |
| MMLU PRO | 87.0 | — | https://huggingface.co/spaces/TIGER-Lab/MMLU-Pro |
| SWE-bench Verified | 76.7 | SWE-bench Verified | https://www.swebench.com/verified.html |
| τ-bench | 78.9 | τ-bench | https://taubench.com/ |
Rankings
Compare
All comparisons →Grok 4 vs GPT-4o (08-06)Grok 4 vs o3Grok 4 vs Claude Sonnet 4.6Grok 4 vs Claude Opus 4.6Grok 4 vs Gemini 3.1 ProGrok 4 vs DeepSeek V4 ProGrok 4 vs DeepSeek R1Grok 4 vs Llama 4 Maverick 17B Instruct FP8Grok 4 vs Qwen3.6-MaxGrok 4 vs GPT-5Grok 4 vs GPT-5.2Grok 4 vs Claude Opus 4.7Grok 4 vs Kimi K2.6Grok 4 vs GPT-5.5Grok 4 vs GPT-5.5 ProGrok 4 vs DeepSeek V4
