Grok 4.5 Takes the Lead in Latest AI Coding Benchmarks

Grok 4.5 is officially making waves after hitting a 29 percent success rate on the SWE Marathon benchmark. This performance puts it ahead of major industry rivals like Claude Opus 4.8 and Fable, signaling a significant jump in capabilities for the AI model.

Beyond raw performance, the pricing strategy is turning heads. Users can now access the model via API at a cost of 2 dollars per million tokens. This aggressive pricing makes it a highly attractive option for developers who need powerful coding tools without the heavy costs associated with other top tier models.

This update is a major milestone for the technology, showing that Grok is becoming a serious player for technical tasks. Traders and tech enthusiasts are watching closely to see if this performance edge leads to wider adoption across the crypto and software development sectors in the coming months.