Grok 4.3 Release on Amazon Bedrock

xAI has announced the general availability of Grok 4.3 on Amazon Bedrock. This release integrates a high-performance frontier model into the AWS ecosystem, providing enterprise developers with a model that emphasizes low hallucination rates and configurable reasoning capabilities.

Enterprise Performance and Benchmarks

Grok 4.3 is positioned as a leading model for building enterprise-ready AI agents, ranking #1 across several key industry benchmarks:

• Hallucination Rates: It achieves the lowest hallucination rate among frontier models, ranking #1 on the Artificial Analysis Omniscience benchmark.
• Tool-Calling: It ranks #1 on the Artificial Analysis Tau2 Telecom benchmark, which measures real-world tool-calling performance specifically for customer support agent scenarios.
• Document Understanding: It ranks #1 on the Vals AI Case Law and Corporate Finance benchmarks, demonstrating high proficiency in complex document understanding tasks.

Technical Capabilities and Context Window

Grok 4.3 features a 1-million-token context window, enabling the processing of extremely long documents and codebases. To optimize performance and cost, the model supports configurable reasoning efforts. Developers can select from four levels of reasoning effort:

• None: Optimized for maximum speed on simple requests.
• Low, Medium, High: Scalable reasoning levels to match the difficulty of the problem being solved.

Pricing and Cost Efficiency

According to xAI, Grok 4.3 delivers 2-10x more intelligence per dollar than other frontier models. The pricing structure for the model on Amazon Bedrock is as follows:

• Input: $1.25 per 1 million tokens
• Output: $2.50 per 1 million tokens

Availability

Grok 4.3 is now available to all developers in supported AWS Regions through the Amazon Bedrock secure inference engine.