SpaceXAI launches Grok 4.5 as new flagship AI model

2026-07-10 · Source: Dataconomy · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Emerging Technologies & Innovation · Depth: Fundamental Awareness, quick

Summary

SpaceXAI has launched Grok 4.5, its new flagship AI model, developed in partnership with AI company Cursor after rebranding from xAI. Grok 4.5 now serves as the default model for SpaceXAI's terminal-based AI coding agent, Grok Build. Trained on tens of thousands of NVIDIA GB300 GPUs using extensive coding, science, engineering, and math datasets, the model is claimed to outperform other leading models in real engineering tasks and excels at generating functional applications from minimal instructions, such as an interactive solar system simulation. Priced at \$2 per million input tokens and \$6 per million output tokens, Grok 4.5 offers significantly lower costs compared to OpenAI's GPT-5.6 Sol (\$5/\$30 per million tokens) while being faster than "flash" models. It also supports tasks in Excel, PowerPoint, and Word, accessible via the SpaceXAI console and Cursor's plans. EU availability is expected mid-July.

Key takeaway

For AI Engineers evaluating large language models for complex engineering or application development, Grok 4.5 presents a compelling option. Its claimed performance on real engineering tasks, combined with a competitive pricing structure of \$2/\$6 per million tokens, suggests a strong value proposition. You should consider benchmarking Grok 4.5 against existing solutions, especially if your workflows involve terminal-based coding agents or require integration with office applications. This could optimize both development efficiency and operational costs.

Key insights

SpaceXAI's Grok 4.5 offers high performance for engineering tasks at competitive costs, integrating with coding agents and office applications.

Principles

• Focus on real engineering tasks.
• Cost-efficiency drives adoption.
• Strategic partnerships enhance development.

Method

The model was trained on tens of thousands of NVIDIA GB300 GPUs using diverse datasets covering coding, science, engineering, and math information.

In practice

• Generate functional apps from prompts.
• Automate Excel, PowerPoint, Word tasks.
• Integrate with terminal-based coding agents.

Topics

• Grok 4.5
• SpaceXAI
• AI Coding Agents
• Large Language Models
• NVIDIA GB300 GPUs
• Token Pricing
• Cursor Partnership

Best for: CTO, VP of Engineering/Data, AI Architect, AI Engineer, Machine Learning Engineer, Director of AI/ML

Related on AIssential

• SpaceXAI’s newest AI model Grok 4.5 dramatically undercuts Anthropic and OpenAI on price — Grok 4.5 offers competitive AI capabilities at half the cost due to superior token efficiency.
• [AINews] SpaceXAI launches Grok 4.5, first Opus-class model post Cursor acquisition — Grok 4.5 delivers near-frontier AI capability with significantly improved cost-efficiency for coding and agent tasks.
• SpaceXAI, Cursor release the strongest Grok yet — Grok 4.5 re-establishes Grok as a competitive frontier model with high performance, speed, and cost efficiency.
• Grok 4.5 IS REALLY GOOD! Opus & GPT Level BUT Faster, Cheaper, & Smarter! (Fully Tested) — Grok 4.5 offers a fast, cost-effective, and token-efficient AI model optimized for coding and agent workflows.
• SpaceX's Grok 4.5 launches at half the price of rivals — here's why that could rattle Anthropic and OpenAI — SpaceX's Grok 4.5 disrupts the AI coding market by prioritizing cost-efficiency and speed over raw benchmark dominance.
• SpaceXAI and Cursor Launch Grok 4.5, Tout Lower Costs Than Rivals — Grok 4.5, a joint SpaceXAI/Cursor model, targets coding and agentic work with lower operational costs.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Dataconomy.