Grok 4.7 arrives on Amazon Bedrock, with AWS agent tests across reasoning levels
SpaceXAI's Grok 4.7 is now available on Amazon Bedrock with configurable reasoning effort, tool calling and structured output, according to the AWS Machine Learning Blog. SpaceXAI says the model uses a larger base model than Grok 4.6 and a longer reinforcement-learning run on a task mix weighted toward multi-hour problems, citing gains such as Terminal-Bench 4.0 rising from 20.3% to 37.6%, plus a 500K-token context window. AWS built a tool-calling agent on three search-requiring tasks with one verifiable answer each and ran it six times per effort level for 72 runs: 67 were correct, with all 18 xhigh runs correct and 15 of 18 at low effort, while mean output tokens were about 1,184 at low versus 9,147 at xhigh.
Because reasoning effort is set per request, teams can choose where each step of an agent sits on the trade-off between thoroughness and token cost.
SpaceXAI’s Grok 4.7 is now available on Amazon Bedrock, which offers a choice of foundation models (FMs) from multiple providers. According to SpaceXAI, Grok 4.7 is trained to work longer on more difficult tasks and knows when to check its own work. That matters most for agents, which carry a task through many steps and must verify what they find along the way.
An agent that works this way has to do more than look things up. It has to form a candidate answer, check it against what it has learned, and refine it until it fits. Grok 4.7 exposes a reasoning effort setting that controls how much deliberation happens before each response, and for agent workloads a practical question is which level to run.
In this post, we build a tool-calling agent on Grok 4.7 and evaluate it on three tasks that require search rather than lookup. The tasks are diagnosing a two-service outage from its alerts, scheduling maintenance under interacting constraints, and choosing an optimal set of migrations under a budget. Each task has exactly one correct answer, verified by brute force. We ran all three at every reasoning effort level, six times each, for 72 agent runs.