Grok 4: pricing, context window and benchmarks
Grok 4 is a frontier model from xAI. It offers a 200,000-token context window, costs $17.90 per million input tokens and $84.75 per million output tokens, and averages 86.5 across the index's benchmark suite.
Pricing
Grok 4 costs $17.90 per million input tokens and $84.75 per million output tokens. A blended 3:1 input-to-output mix works out to about $34.61 per million tokens. Cached input is listed at $2.45 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $17.90 | per 1M tokens |
| Output | $84.75 | per 1M tokens |
| Cached input | $2.45 | per 1M tokens |
| Blended 3:1 | $34.61 | estimate per 1M tokens |
| 1M in + 1M out | $102.65 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 87.8 | 20 | 97th | |
| GPQA Diamond | 82.6 | 24 | 96th | |
| SWE-bench Verified | 77.9 | 17 | 97th | |
| HumanEval | 92.0 | 22 | 96th | |
| MATH-500 | 90.8 | 37 | 94th | |
| AIME 2025 | 82.2 | 63 | 90th | |
| LiveCodeBench | 82.2 | 23 | 96th | |
| IFEval | 91.6 | 43 | 93th | |
| TruthfulQA | 87.6 | 28 | 95th | |
| HellaSwag | 93.9 | 13 | 98th | |
| MMMU | 82.9 | 24 | 96th | |
| Arena Elo | 1457.8 | 58 | 90th |
What Grok 4 is good for
Grok 4 sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.90 per million input tokens and a 200,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.
Strengths
- Top-tier average benchmark score (86.5), placing it #26 of 600.
- Large 200,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $17.90 per million tokens.
- Cost can climb quickly on long outputs and high-volume traffic.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Claude Sonnet 4.5 Pro | Anthropic | 1000000 | $18.73 | 86.7 |
| Grok 4 Turbo | xAI | 1000000 | $19.21 | 86.7 |
| Grok 4 Max | xAI | 2000000 | $19.70 | 86.6 |
| o4 Turbo | OpenAI | 2000000 | $19.34 | 86.6 |
| Grok 4 | xAI | 200000 | $17.90 | 86.5 |
| Gemini 3 Pro Turbo | Google DeepMind | 400000 | $19.62 | 86.4 |
FAQ
How much does Grok 4 cost?
Grok 4 is listed at $17.90 per million input tokens and $84.75 per million output tokens. Cached input is $2.45 per million tokens. A blended 3:1 input-to-output mix is about $34.61 per million tokens.
What is the context window of Grok 4?
Grok 4 has a context window of 200,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is Grok 4 open weights?
No. Grok 4 is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Grok 4 score on benchmarks?
Grok 4 averages 86.5 across the scored benchmarks in this index, ranking #26 of 600 models. Its strongest relative result is HellaSwag at 93.9.
What is Grok 4 best for?
Grok 4 sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.90 per million input tokens and a 200,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.