SpaceXAI has unveiled Grok 4.5, the company\u2019s latest large language model and its first major AI release since going public. The launch arrives at a time when competition among leading AI developers is increasingly centred not only on model capability but also on the cost and efficiency of running advanced systems at scale.<\/p>\n
According to the company, Grok 4.5 is designed to handle a broad range of day-to-day professional tasks, including software development, application building, research, writing and office work. Rather than positioning the model around a single speciality, SpaceXAI is presenting it as an all-purpose system capable of supporting routine knowledge work across industries.<\/p>\n
STORY CONTINUES BELOW THIS AD<\/p>\n
Announcing Grok 4.5, our first model trained specifically for coding and agents. It was trained with Cursor and offers frontier intelligence at leading speeds and cost efficiency.https:\/\/t.co\/i8HpU7w64k<\/a> pic.twitter.com\/oBjGtTsoNc<\/a><\/p>\n \u2014 SpaceXAI (@SpaceXAI) July 8, 2026<\/a>SpaceXAI launches Grok 4.5 with focus on speed and pricing<\/p>\n One of the company\u2019s biggest claims is that Grok 4.5 delivers significantly better token efficiency than competing models. SpaceXAI says the new model uses tokens twice as efficiently as other leading AI systems, a metric that could translate into lower operating costs if the performance holds up in real-world deployments. Token usage has become an increasingly important consideration for businesses adopting generative AI, as inference costs remain one of the biggest barriers to large-scale implementation.<\/p>\n Grok 4.5 is efficient and fast. It uses half as many tokens per task, delivers higher tokens per second, and costs less than half as much as comparable models, at $2\/M input and $6\/M output tokens.<\/p>\n \u2014 SpaceXAI (@SpaceXAI) July 8, 2026<\/a><\/p>\n Alongside the announcement, SpaceXAI released benchmark results intended to demonstrate Grok 4.5\u2019s competitiveness against other frontier AI models. While the published scores place the model among the industry\u2019s strongest offerings, the company stopped short of claiming outright leadership across every benchmark.<\/p>\n Musk compares Grok to Anthropic\u2019s Opus as AI competition intensifies<\/p>\n Founder Elon Musk also promoted the launch on X, the social media platform owned by SpaceXAI, highlighting both the model\u2019s performance and economics.<\/p>\n \u201cBased on strong positive feedback from customers in our beta test program, @SpaceXAI will make Grok 4.5 available to the public tomorrow. It is an Opus-class model, but faster, more token-efficient and lower cost,\u201d Musk wrote.<\/p>\n Our internal assessment is that Grok 4.5 is roughly comparable to Opus 4.7, but much faster. The combination of capability, faster speed and lower cost is what makes it competitive.<\/p>\n We are closing the loop on real-world usefulness, not benchmarks. Hardcore engineers at Tesla &\u2026 https:\/\/t.co\/FqoJzHkrRY<\/a><\/p>\n