SpaceXAI has unveiled Grok 4.5, its latest artificial intelligence model, positioned as a practical tool for wide-ranging knowledge work including software development, administrative tasks, research, and writing. The company emphasises the model's efficiency advantage, claiming it processes tokens at twice the rate of leading competitors, a feature designed to address growing cost concerns among AI users. Benchmarks released alongside the announcement demonstrate Grok's competitive standing, though the model does not consistently rank as best-in-class across all metrics.

The release comes as intense competition in the AI industry drives down pricing and raises performance expectations. SpaceXAI's pricing—£2 per million input tokens and £6 per million output tokens—undercuts Anthropic's Opus 4.7 significantly, which costs £5 input and £25 output. Founder Elon Musk characterised Grok 4.5 as comparable to Opus 4.7 in capability but with faster execution and lower operational costs, positioning SpaceXAI's new offering as a value-focused alternative. This release occurs during an especially competitive week, with OpenAI preparing to launch GPT 5.6, its most advanced model yet.