Grok
SpaceXAI's models, strong on long-running engineering tasks.
Grok is SpaceXAI's model family, sold through its own API rather than published as weights. The recent versions are pitched at coding and agentic work — long software-engineering runs where the model checks its own output — and take text, images and files with a 500K-token context. Closed weights and paid only, so treat it as a frontier option you rent, not one you host. The Grok chat app is listed separately under Chats.
At a glance — Grok 4.7
- Cost
- $1.60 in · $4.80 out per 1M tokensMid-range
- Memory
- 500K tokens — about 750 pages
- Reads
- Text · Images · Files
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Overall: better than 89%
- Reliability
- 99.93% uptime, last 24h
- Released
- Sep 21, 2026
Advanced details
- API model ID
x-ai/grok-4.7- Max output
- 450K tokens
- Cached input
- $0.40 per 1M
- Prompts over 200K
- $3.20 in · $9.60 out per 1M
- Tokenizer
- Grok
- Overall index
- 46.4
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| xAI · zdr | $1.60 | $4.80 | $0.40 | 450K | 99.93% |
| xAI | $1.60 | $4.80 | $0.40 | 450K | 99.40% |
| xAI · zdr | $3.20 | $9.60 | $0.80 | 450K | 99.17% |
| xAI · priority | $3.20 | $9.60 | $0.80 | 450K | 97.09% |
Supported parameters
- include_reasoning
- logprobs
- max_tokens
- reasoning
- reasoning_effort
- response_format
- seed
- structured_outputs
- temperature
- tool_choice
- tools
- top_logprobs
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Paid
- Category
- Models
- Best for
- Intermediate
- Made by
- SpaceXAI
- Weights
- Closed, API only
- Good for
- Long software-engineering tasks