xAI
Grok 4.7
Grok 4.7 scores 1637 on WebDev Arena — rank 11 of 29 models we track with a published score and costs $3 per 1M blended tokens on the API — the best-scoring model at that price is Qwen3.8-Max (1674).
API price
$3
per 1M blended tokens
WebDev Arena
1637
rank 11 of 29 tracked models
Benchmark scores
| Benchmark | Score | Rank | Source | Published |
|---|---|---|---|---|
| WebDev Arena | 1637xHigh | #11 of 29 | LMArena | 2026-10-19 |
| FrontierCode | 47.6%unknown | #8 of 25 | Epoch AI Benchmarking Hub | 2026-09-21 |
| FrontierSWE | 29.5%xhigh | #8 of 12 | Epoch AI Benchmarking Hub | 2026-09-21 |
| CursorBench | 46.3%xhigh | #6 of 10 | Epoch AI Benchmarking Hub | 2026-09-21 |
The rank is among the 29 models we track with a published score for that benchmark — not the full leaderboard of the source, which lists many models we do not track. Where a source publishes several reasoning-effort settings, we show the model's best listed result.
API price
- Input
- $2 per 1M tokens
- Output
- $6 per 1M tokens
- Blended (3 input : 1 output)
- $3 per 1M tokens
Context window is 500k tokens.
Checked 2026-10-21 · source
Where it sits
WebDev Arena score against API price, with every model we can price and score on the same benchmark. Up and to the left is better.
- Value frontier
- No — a cheaper model scores higher
- Best model at $3 or less
- Qwen3.8-Max1674 · $3
- Cheaper and scoring higher
- DeepSeek V4.1 Flash1664 · $0.52
- Cheaper models on the frontier
- Claude Haiku 5.51586 · $0.20, GLM-5.3-Flash1607 · $0.24, DeepSeek V4.1 Flash1664 · $0.52
Plans that include Grok 4.7
- Copilot Pro+
GitHub
- Copilot Max
GitHub
Other xAI models
Common questions
- How much does Grok 4.7 cost?
- $2 per 1M input tokens and $6 per 1M output tokens, which is $3 on a 3:1 input:output mix. The price was checked on 2026-10-21.
- Which subscription plans include Grok 4.7?
- 2 of the plans we track list it by name: Copilot Pro+ ($39/mo), Copilot Max ($100/mo).
- How does Grok 4.7 rank on WebDev Arena?
- It ranks 11 of 29 models we track with a published WebDev Arena score, at 1637. The rank is among the models on this site, not the full LMArena leaderboard.
- Is Grok 4.7 the best value for coding?
- No. 1 tracked model score higher on WebDev Arena and cost less. The best-scoring model at $3 or less is Qwen3.8-Max.
- What is the cheapest way to use Grok 4.7?
- Copilot Pro+ is the cheapest subscription we track that lists it, at $39/month. Per-token API pricing is listed separately above.