Grok 4.7
grok
xAI's frontier model for long-running agents, coding, knowledge work, and visual projects
ReasoningTool CallingAttachmentsStructured Output
Context Window
500K
Max Output
500K
Temperature
Yes
Open Weights
No
Knowledge Cutoff
2026-05
Released
2026-09-21
Last Updated
2026-09-21
Modalities
Input:TextImagePDF
→Output:Text
Benchmarks
| Benchmark | Score | Metric |
|---|---|---|
| CursorBench | 46.3 | score |
| DeepSWE | 71 | score |
| EEBench | 64 | score |
| AA-Briefcase | 1657 | Elo |
| Terminal-Bench | 37.6 | score |
| Harvey Legal Agent Benchmark | 19.6 | score |
| HealthBench | 56.7 | score |
| GDPval | 1695 | Elo |
Available Providers (5)
| Provider | Input /1M | Output /1M | Cache Read /1M | Cache Write /1M | Reasoning | Status |
|---|---|---|---|---|---|---|
 Cloudflare AI Gateway | Rp 35.761 | Rp 107.281 | Rp 8.941 | — | effort | — |
 Merge Gateway | Rp 35.761 | Rp 107.281 | Rp 8.941 | — | effort | — |
 Eden AI | Rp 35.761 | Rp 107.281 | Rp 8.941 | — | effort | — |
 Tempr Gateway | Rp 35.761 | Rp 107.281 | Rp 8.941 | — | effort | — |
 Vertex | Rp 35.761 | Rp 107.281 | Rp 8.941 | — | effort | — |