SpacexAI Releases Grok 4.5, Claiming Opus-Class Performance at Lower Cost as OpenAI Prepares Competing Launch

SpaceXAI released its latest AI model, Grok 4.5, its first launch since going public several weeks ago. The company described the model as capable of handling coding, app-building, office work, research and writing, claiming twice the token efficiency of competing models. Benchmark results released Wednesday showed Grok 4.5 performing competitively against rival models, though slightly behind top performers in some categories.

Founder Elon Musk compared the model to Anthropic’s Opus, writing on X that Grok 4.5 would become publicly available following positive feedback from beta testers, describing it as an “Opus-class” model that is faster and more cost-efficient. Musk later said the company’s internal assessment placed Grok 4.5 roughly on par with Opus 4.7, though notably faster. SpaceXAI priced the model at $2 per million input tokens and $6 per million output tokens, compared to Opus 4.7’s $5 and $25 respectively, and OpenAI’s tiered pricing ranging from $1 to $5 per million input tokens depending on model tier.

The release comes as OpenAI prepares to launch GPT-5.6, which it has called its strongest model yet, after the release had previously faced limitations tied to security concerns raised by the Trump administration.