grok-4-7

grok-4-7 is a frontier-tier model from xAI in the Grok 4 family, priced at $2.00 per million input tokens and $6.00 per million output tokens. Prompt caching reduces repeated input to $0.50/1M. It supports a 256K-token context window, vision inputs, function calling, and streaming. grok-4-7 is a proprietary model available exclusively through xAI's API.

grok-4-7 is priced at $2.00 per million input tokens and $6.00 per million output tokens, with prompt caching at $0.50/1M. The 3:1 output-to-input ratio is lower than many frontier models, making it particularly cost-effective for output-heavy workloads. At 1,000 input + 500 output tokens per call, the cost is ($2.00 × 1,000 + $6.00 × 500) ÷ 1,000,000 = $0.005 per call, or approximately $5.00 per 1,000 API calls — competitive with advanced-tier models from other providers. Prompt caching at $0.50/1M provides savings for workloads with long repeated system prompts.

grok-4-7's 256K-token context window is in the top tier for production models, enabling full-document analysis, large codebase understanding, and long multi-turn conversations without chunking. Maximum output is 131,072 tokens — sufficient for long-form content, detailed code files, and extended reports. Median latency is approximately 900ms TTFT (Estimate). Throughput is approximately 90 tokens per second (Estimate). As a standard (non-reasoning) model, it delivers faster responses than reasoning-tier models.

grok-4-7 is best suited to reasoning, coding, and chat workloads where frontier capability is required at a competitive price point. Its $2/$6 pricing with a 3:1 output ratio makes it particularly attractive for output-heavy workloads — long-form generation, detailed code files, and extended reports benefit from the lower output-to-input price ratio compared to models with 5:1 ratios. For workloads requiring extended chain-of-thought reasoning, grok-4-0709 ($3/$15) offers deeper reasoning capability at higher cost.

grok-4-7 sits in the Grok 4 family alongside grok-4-0709 ($3/$15, reasoning model). At $2/$6, it offers frontier-tier capability at a lower price point than grok-4-0709, with the trade-off of standard (non-reasoning) processing. Its intelligence score of approximately 90/100 reflects strong frontier-tier performance. xAI's API is newer and has a smaller ecosystem than established providers — teams should evaluate production reliability and tooling support accordingly. Main competitors are frontier-tier models from OpenAI, Anthropic, and Google at similar price points.