Grok 4.7
xAI · Released 21 Sep 2026 · grok-4.7
$2.00
Input price / 1M tokens
#36 of 61, cheapest first
$6.00
Output price / 1M tokens
#32 of 61, cheapest first
500K
Context window
#51 of 71
79
Output tokens / second
#41 of 61
71%
DeepSWE v1.1
#7 of 17
37.6%
Terminal-Bench 4.0
#8 of 10
Summary
Grok 4.7 is a proprietary model from xAI, released on 21 Sep 2026.
At $2.00 input and $6.00 output per million tokens, it is #36 of 61 on input price, cheapest first.
Its 500K-token context window ranks #51 of 71.
Measured output speed is 79 tokens per second, #41 of 61.
xAI's newest flagship for coding, agentic tasks and knowledge work, with configurable reasoning effort (low to xhigh). xAI recommends it for everything except image, video and voice.
Benchmarks · 6 reported
| Benchmark | Score | Bar | Rank | Source |
|---|---|---|---|---|
| CursorBench 4.0xhigh effort | 46.3% | – | xAI launch postSelf-reported | |
| DeepSWE v1.1high effort | 71% | #7 of 17 | xAI launch postSelf-reported | |
| Terminal-Bench 4.0xhigh effort | 37.6% | #8 of 10 | xAI launch postSelf-reported | |
| EEBenchxhigh effort | 64% | – | xAI launch postSelf-reported | |
| HealthBench Professionalxhigh effort | 56.7% | – | xAI launch postSelf-reported | |
| GDPvalxhigh effort | 1,695 | Elo-style | – | xAI launch postSelf-reported |
Self-reported means the lab ran the test itself. A rank appears only where several labs report the same version of a benchmark.
Similar models
Models with the most benchmarks in common with Grok 4.7, and the closest scores.
DeepSWE v1.1
#7 of 17Long coding tasks in real repos
- Grok 4.771
- Gemini 3.8 Flash73.7
- GPT-6 Astra74.1
- GPT-5.6 Terra69.6
- GPT-5.6 Sol72.7
- GPT-6 Sol68.8
Terminal-Bench 4.0
#8 of 10Long jobs in a real terminal
- Grok 4.737.6
- Gemini 3.8 Flash19.1
- GPT-6 Astra57.9
Compare Grok 4.7 side by side
Pre-filled with the two closest models. Swap any of them.
Sources
- price, cached price, long-context tier, batch and priority rules: docs.x.ai/developers/pricing, checked 2 Oct 2026
- context, modalities, batch not supported, rate limits: docs.x.ai/developers/models/grok-4.7, checked 2 Oct 2026
- knowledge cutoff, output limit, reasoning levels: docs.x.ai/developers/grok-4-7, checked 2 Oct 2026
- release date and benchmarks: x.ai/news/grok-4-7, checked 2 Oct 2026
- release notes: docs.x.ai/developers/release-notes, checked 2 Oct 2026
- free Playground, no free API usage: x.ai/api, checked 2 Oct 2026
- output speed: artificialanalysis.ai/models/grok-4-7, checked 2 Oct 2026
Building on Grok 4.7?
Get the architecture right before the bill arrives.
We size caching, routing and fallbacks for your workload on a free call, and tell you where a cheaper model is good enough.