SpaceXAI Introduces Grok 4.5, Advanced AI Model for Coding and Agentic Tasks

SpaceXAI, the AI company formerly known as xAI before this month's rebrand that followed February's $1.25 trillion merger with SpaceX, has released Grok 4.5, its most capable model yet for coding, agentic tasks, and knowledge work. The model was trained alongside Cursor, the AI coding editor, and targets real-world engineering work spanning Rust and C/C++ tasks through end-to-end app development from prompt to production.

SpaceXAI is pricing Grok 4.5 at $2 per million input tokens and $6 per million output tokens, undercutting several rival frontier models, and says the model runs at 80 tokens per second while using roughly 4.2 times fewer output tokens than Opus 4.8 (max) on the SWE-Bench Pro benchmark. On coding benchmarks, Grok 4.5 posted a 64.7% resolve rate on SWE-Bench Pro and 83.3% on Terminal Bench 2.1, trailing the top-scoring models in each test but at a fraction of the cost. Elon Musk called it an Opus-class model, but faster, more token-efficient, and lower cost.

Grok 4.5 is immediately available through Cursor on all plans, SpaceXAI's own Grok Build coding agent, and the company developer console, though the rollout excludes the EU until mid-July. The release is SpaceXAI's first major model launch since the SpaceX-xAI combination, and it lands as SpaceX reportedly prepares a blockbuster IPO later this year, raising the stakes for how enterprise and developer adoption of Grok 4.5 will be read as a proof point for the newly merged company's AI ambitions.

Related Stocks

• SPCX