# Grok 4.5

> Grok 4.5 is available on Artifio.ai — a unified AI workspace with a single pay-as-you-go wallet.

- Canonical page: https://artifio.ai/models/grok-4-5
- Run it: https://artifio.ai/create/grok-4-5
- Modality: prompt to text
- Model brand: xAI
- Pricing: from about $0.150 per generation (exact cost shown in the workspace before you confirm; varies with selected options)
- Typical generation time: ~25s (catalogue estimate)
- Full pricing table: https://artifio.ai/pricing/models

## About Grok 4.5

Grok 4.5 is the frontier text model from xAI (which now operates under the SpaceXAI name), released on July 8, 2026. The company built it for coding, agentic tasks, and knowledge work, training it in its Memphis data centers on new datasets spanning science, engineering, and math. Elon Musk pitched it as an Opus-class model that is faster, more token-efficient, and lower cost than Anthropic's flagship, and independent benchmarking broadly backed the framing while stopping short of best-in-class.

It is the default model of the Grok Build coding agent and ships across Cursor, the Word, PowerPoint, and Excel add-ins, and major model gateways. On artifio.ai it runs in the same workspace as 100+ other models, billed per generation from one wallet.

## Features

- 500,000-token context window, with a higher rate applied to requests past 200K tokens.
- Configurable reasoning effort at low, medium, or high, defaulting to high.
- Text and image input with text output.
- Function calling, structured outputs, and server-side web search, X search, and code execution tools.
- Prompt caching with cached input tokens billed at a quarter of the standard rate.
- Knowledge cutoff of February 1, 2026.
- Available through the xAI API, Grok Build, Cursor, Office add-ins, and gateways including OpenRouter, Vercel, Cloudflare, Snowflake, and Databricks.

## What people use it for

- Agentic coding: it is the default engine of Grok Build and was tuned on real editor session data.
- Long-running legal, finance, and research tasks that chain many tool calls.
- Office and clerical work such as drafting, analysis, and document preparation inside the Microsoft add-ins.
- Cost-sensitive deployments that want near-frontier reasoning without frontier per-token prices.

## Reported strengths

- Scored 54 on the Artificial Analysis Intelligence Index, fourth overall behind only Fable 5, GPT-5.5, and Opus 4.8, and 16 points above Grok 4.3.
- Scored 76 on the Artificial Analysis Coding Agent Index in the Grok Build harness, on par with GPT-5.5 in Codex.
- Exceptional token efficiency: about 1.9M tokens per coding task versus 6.2M for GPT-5.5 and 7.2M for Fable 5 in the same evaluation.
- Priced well below Opus-class competitors per token while landing in their benchmark neighbourhood.

## Reported limitations

- Artificial Analysis measured its hallucination rate rising from 25 percent to 54 percent versus Grok 4.3, a calibration trade-off that came with the accuracy gains.
- The 500K context window is half of Grok 4.3's 1M, a step backward for very long documents.
- Not available to EU users in the xAI API console at launch.
- Costs more per token than Grok 4.3, and requests past 200K context tokens are billed at a higher rate again.

## Sources

1. grok-4.5 overview (xAI Docs): https://docs.x.ai/developers/grok-4-5
2. grok-4.5 model page (xAI Docs): https://docs.x.ai/developers/models/grok-4.5
3. SpaceXAI releases Grok 4.5, which Elon describes as an 'Opus-class model' (TechCrunch): https://techcrunch.com/2026/07/08/spacexai-releases-grok-4-5-which-elon-describes-as-an-opus-class-model/
4. Grok 4.5 brings SpaceXAI to the intelligence frontier (Artificial Analysis): https://artificialanalysis.ai/articles/grok-4-5-brings-spacexai-to-the-the-intelligence-frontier

---

About Artifio.ai: 100+ AI image, video, and audio models in one workspace with a single shared wallet — no per-model subscriptions. This is the markdown rendition of https://artifio.ai/models/grok-4-5 for AI assistants and crawlers; the HTML page is canonical.