Grok 4.7 Now Available on Amazon Bedrock
xAI’s Grok 4.7 model, optimized for coding, long‑running agents and knowledge work, is now offered through Amazon Bedrock with a 500K token context and configurable reasoning effort.
Az eredeti forrás közzétételi dátuma: 2026. szeptember 28.
xAI’s Grok 4.7 model is now accessible via Amazon Bedrock, targeting coding, long‑running agents and professional knowledge tasks. It provides a 500 K token context window and four reasoning‑effort settings (low, medium, high, xhigh). The model emphasizes endurance, self‑verification of its output, and improved use of long contexts, and it includes a new safeguard stack that strengthens refusal handling and jailbreak resistance.
On Bedrock, Grok 4.7 is served through the bedrock‑runtime endpoint and supports the Responses, Chat Completions, InvokeModel and Converse APIs with text and image inputs. It can be invoked via two cross‑Region inference profiles—global.xai.grok-4.7 for cost‑optimized routing and us.xai.grok-4.7 for US‑data‑residency or latency‑sensitive workloads. Independent evaluation by Artificial Analysis shows performance gains on long‑horizon agentic work and coding agents, albeit with higher token usage, making the model a useful study case for learners building large‑context, self‑checking AI agents.