SpaceXAI Releases Grok 4.5, a Cursor-Trained Model for Coding, Agentic Tasks, and Knowledge Work at $2/M Input

SpaceXAI just released Grok 4.5 . The company calls it its smartest model to date. It targets coding, agentic tasks, and knowledge work. SpaceXAI says Grok 4.5 was trained alongside Cursor, an AI coding editor.

Grok 4.5 is a general-purpose model tuned for real engineering work. SpaceXAI trained it on datasets spanning coding, science, engineering, and math. The research team describes its reasoning as both intelligent and efficient. It scored #1 on Harvey’s Legal Agent Benchmark, which SpaceXAI cites as office-work strength.

Training ran across tens of thousands of NVIDIA GB300 GPUs. SpaceXAI used training and stability techniques designed for large-scale runs. Beyond raw token volume, the team invested in data filtering and curation. This included deduplication, quality scoring, and domain-focused selection.

SpaceXAI team then scaled reinforcement learning with a focus on per-token intelligence. RL covered hundreds of thousands of tasks. Most centered on multi-step software engineering and other technical work. Grading combined automated and model-based methods. The stack supports highly asynchronous training. Agentic rollouts can run for many hours while learning continues.

SpaceXAI team published scores across four coding benchmarks. Competitor figures come from published system cards or leaderboards. SpaceXAI’s prose says Grok 4.5 exceeds comparable leading models. Its own chart is more mixed. Fable (max) posts the top score on all four benchmarks. Grok 4.5 stays closest on Terminal Bench 2.1.

Quick reference: “pass@1” counts only first-attempt passes; “resolve rate” is the share of tasks fixed.

Grok 4.5 is served at 80 TPS . SpaceXAI reports roughly twice the token efficiency of leading models. On SWE Bench Pro, Grok 4.5 resolved tasks with 15,954 output tokens on average. SpaceXAI reports Opus 4.8 (max) used 67,020 on the same benchmark. That is about 4.2× fewer output tokens. Fewer output tokens usually means lower output cost and latency per task.

Grok 4.5 costs $2 per million input tokens and $6 per million output tokens . SpaceXAI says it solves tasks in under half the number of steps. Confirm current pricing in the SpaceXAI console before budgeting.

Grok 4.5 is available in Grok Build, in Cursor on all plans, and from the SpaceXAI console. Grab an API key and call the responses endpoint. The model ID is grok-4.5 .

To use Grok Build from the terminal, install the CLI: