Grok 3 Turbo: pricing, context window and benchmarks
Grok 3 Turbo is a multimodal model from xAI. It offers a 128,000-token context window, costs $15.97 per million input tokens and $70.13 per million output tokens, and averages 83.1 across the index's benchmark suite.
Pricing
Grok 3 Turbo costs $15.97 per million input tokens and $70.13 per million output tokens. A blended 3:1 input-to-output mix works out to about $29.51 per million tokens. Cached input is listed at $1.48 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $15.97 | per 1M tokens |
| Output | $70.13 | per 1M tokens |
| Cached input | $1.48 | per 1M tokens |
| Blended 3:1 | $29.51 | estimate per 1M tokens |
| 1M in + 1M out | $86.10 | 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 | 83.7 | 88 | 85th | |
| GPQA Diamond | 79.1 | 71 | 88th | |
| SWE-bench Verified | 71.4 | 88 | 85th | |
| HumanEval | 86.8 | 93 | 85th | |
| MATH-500 | 85.6 | 99 | 84th | |
| AIME 2025 | 83.1 | 53 | 91th | |
| LiveCodeBench | 76.4 | 78 | 87th | |
| IFEval | 90.7 | 61 | 90th | |
| TruthfulQA | 86.0 | 51 | 92th | |
| HellaSwag | 90.2 | 81 | 87th | |
| MMMU | 81.3 | 41 | 93th | |
| Arena Elo | 1438.4 | 101 | 83th |
What Grok 3 Turbo is good for
Grok 3 Turbo is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $15.97 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 128,000-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 $15.97 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Claude 3.7 Sonnet Pro | Anthropic | 128000 | $16.07 | 83.3 |
| DeepSeek-R1 Instruct | DeepSeek | 1000000 | $5.90 | 83.1 |
| Grok 3 Turbo | xAI | 128000 | $15.97 | 83.1 |
| Kimi K2 | Moonshot AI | 128000 | $5.85 | 83.0 |
| GPT-4.1 Turbo | OpenAI | 1000000 | $16.12 | 82.9 |
| Grok 3 Reasoning | xAI | 200000 | $16.07 | 82.9 |
FAQ
How much does Grok 3 Turbo cost?
Grok 3 Turbo is listed at $15.97 per million input tokens and $70.13 per million output tokens. Cached input is $1.48 per million tokens. A blended 3:1 input-to-output mix is about $29.51 per million tokens.
What is the context window of Grok 3 Turbo?
Grok 3 Turbo has a context window of 128,000 tokens and can emit up to 65,536 output tokens per request in this index.
Is Grok 3 Turbo open weights?
No. Grok 3 Turbo is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Grok 3 Turbo score on benchmarks?
Grok 3 Turbo averages 83.1 across the scored benchmarks in this index, ranking #73 of 600 models. Its strongest relative result is Arena Elo at 1438.4.
What is Grok 3 Turbo best for?
Grok 3 Turbo is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $15.97 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.