Spacexai launches grok 4.7, its most advanced coding and knowledge processing model

SpaceXAI has launched Grok 4.7, its most advanced coding and knowledge processing model to date. The model is designed for programming and knowledge work, offering twice the speed of similar models at half the cost. Grok 4.7 features a larger base model and extended reinforcement learning training, enhancing its ability to handle complex tasks. It also improves self-validation and long-context management. In CursorBench 4.0, Grok 4.7 performs competitively in long-running programming tasks. The model also shows improvements in GDPval and AA Briefcase benchmarks, reaching performance levels comparable to other leading models. SpaceXAI has implemented a new security framework, achieving high levels of safety in rejecting harmful prompts and resisting jailbreaking attacks. Grok 4.7 is available through Cursor and Grok Build, with API access and support for third-party tools. Pricing starts at $2 per million tokens for input and $6 for output, with a faster version available at double the price.