Summary
SpaceXAI released Grok 4.5, its first model since going public and acquiring Cursor, pitched as a coding and agentic-work model rather than a consumer chatbot. Priced at $2 per million input and $6 per million output tokens, it undercuts flagship rivals and claims roughly 4.2x fewer output tokens than Claude Opus 4.8 on SWE Bench Pro.
What changed
On July 8, 2026 SpaceXAI launched Grok 4.5 at $2/M input and $6/M output tokens, available in Grok Build, in Cursor on all plans, and via the SpaceXAI console (not yet in the EU). Trained alongside Cursor, it targets engineering, agentic, and knowledge work with high token efficiency.
Why it matters
Aggressive pricing roughly 60% below Claude Opus 4.8 on input and 76% below on output, combined with strong token efficiency on coding benchmarks, pressures Anthropic and OpenAI on the cost of agentic coding workloads. Owning Cursor gives SpaceXAI a large distribution surface to push the model directly into developer workflows.
Evidence excerpt
SpaceXAI launched Grok 4.5 on July 8, 2026, pricing the model at $2 per million input tokens and $6 per million output tokens, far below Anthropic and OpenAI's flagship rates; about 4.2x fewer output tokens than Opus 4.8 on SWE Bench Pro.