xAI Launches Grok 4.5: Elon Musk Calls It an 'Opus-Class' AI Model at a Fraction of the Cost

xAI Unveils Grok 4.5, Its Most Capable Model Since Going Public

Elon Musk's artificial intelligence company, xAI, has officially launched Grok 4.5, its latest large language model — and the first major release since the company made its public debut on the stock market just weeks ago. The company is positioning Grok 4.5 as a high-performance yet cost-effective alternative to the most powerful AI models currently available.

What Grok 4.5 Can Do

According to a blog post published Wednesday, xAI describes Grok 4.5 as a robust, all-purpose AI workhorse built to handle the wide range of tasks that the industry has been racing to automate. These include software development and app creation, administrative and clerical work, academic and professional research, content writing, and general knowledge-based tasks.

The company isn't just competing on capability — it's making a strong case on economics as well. xAI claims that Grok 4.5 delivers twice the token efficiency compared to other leading models on the market. For businesses and developers who have been increasingly concerned about the rising cost of AI usage, this could be a compelling advantage if the claims hold up in real-world applications.

Benchmark Performance and Competitive Standing

Alongside the announcement, xAI released benchmark data suggesting that Grok 4.5 performs competitively against top-tier models from rival companies — though it falls just short of claiming the top spot in every category. The results paint a picture of a model that punches well above its price point, even if it isn't definitively best-in-class across the board.

Musk Compares Grok 4.5 to Anthropic's Opus

Posting on X — the social media platform he owns, which operates as a subsidiary of xAI — Musk drew a direct comparison between Grok 4.5 and Opus, Anthropic's flagship model engineered for handling intensive and highly complex tasks.

"Based on strong positive feedback from customers in our beta test program, @xAI will make Grok 4.5 available to the public tomorrow. It is an Opus-class model, but faster, more token-efficient and lower cost," Musk wrote.

He followed up with a more specific assessment:

"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."

How Grok 4.5 Stacks Up on Pricing

One of the most striking aspects of Grok 4.5's launch is its pricing structure. xAI has set the cost at $2 per million input tokens and $6 per million output tokens — numbers that look particularly attractive when placed alongside the competition.

For reference:

• Anthropic's Opus 4.7 is priced at $5 per million input tokens and $25 per million output tokens.
• OpenAI's Sol (its premium tier) runs $5 per million input tokens and $30 per million output tokens.
• OpenAI's Luna (its budget tier) costs $1 per million input tokens and $6 per million output tokens.

If Grok 4.5's performance truly mirrors its pricing promise, xAI could shake up the current AI model pricing landscape in a significant way.

A Big Week for AI Releases

The timing of Grok 4.5's launch is notable, as it coincides with what is shaping up to be a landmark week for artificial intelligence announcements. OpenAI is reportedly set to release GPT-5.6 on Thursday — described by the company as its "strongest model yet." The rollout of this model had previously been delayed due to security-related concerns raised by the Trump administration, adding an extra layer of anticipation to its imminent debut.

With multiple major AI players releasing new models in rapid succession, consumers, developers, and enterprises now have more options — and more comparisons to make — than ever before.