Grok on GlobalGPT
Bring a difficult question, a writing brief or a code snippet to Grok 4.7. Compare the result with Grok 4.6, or take a visual idea into the separate Grok Imagine video workspace.
Curated by GlobalGPT Review Team
Choose Your Starting PointGrok 4.7
Grok Chat and Imagine, Together in GlobalGPT.
Use Grok for text work and Grok Imagine for video within GlobalGPT’s multi-model subscription. The official Grok app, Grok Build, X subscriptions and the xAI API have their own tools, access rules and billing.
Selected Milestones in the Grok Family.
Selected milestones from official xAI documentation. Earlier dates refer to API releases, not GlobalGPT launch dates; Imagine is a separate media family.
Choose Chat for Answers, Imagine for Video.
Start with Grok 4.7 for demanding text work, compare it with 4.6, or open Grok Imagine Video 1.5 for a video project.
Grok 4.7
Work through a coding question, conflicting source excerpts or a detailed brief. Define the expected output and verify important claims and code before using them.
Grok 4.6
Use the same writing brief, source excerpts or code sample to compare with Grok 4.7. Check facts, preserved constraints and executable results yourself.
Grok Imagine Video 1.5
Use the dedicated video workspace to animate a starting frame and direct movement and sound. Check the available controls before generating.
Previous Grok ModelsKeep older entries for comparison. Their route and access rules may change; start with a current model for a new task.
Grok 4.3Previous chat option · Chat model
Compare writing, summaries and everyday questions with the same brief. Choose based on the result and your account’s usage rules./home/grok-4-3?inviter=hub_research_grok&login=1
Compare Models, Access and Costs
Choose by workload, then review official subscriptions, developer API prices and published benchmark conditions.
Grok API Token Prices
| Model / API | Input / 1M tokens | Cached input / 1M | Output / 1M | Prompt size |
|---|---|---|---|---|
| Grok 4.7xAI xAI official documentation ↗Last verified: | $2.00 | $0.50 | $6.00 | Under 200K tokens |
| Grok 4.7 long contextxAI xAI official documentation ↗Last verified: | $4.00 | $1.00 | $12.00 | 200K tokens or more |
| Grok 4.6xAI xAI official documentation ↗Last verified: | $2.00 | $0.50 | $6.00 | Under 200K tokens |
| Grok 4.6 long contextxAI xAI official documentation ↗Last verified: | $4.00 | $1.00 | $12.00 | 200K tokens or more |
USD direct xAI text-token pricing. Tool calls, regional processing and media generation are separate. Grok 4.7 Fast pricing is documented for Cursor and Grok Build, not the public API. These rates are not GlobalGPT subscription prices.
Official Plans: The X Subscription
| Plan | USD monthly | USD annual | Included shared usage |
|---|---|---|---|
| FreexAI xAI official documentation ↗Last verified: | $0 | $0 | Limited |
| PlusxAI xAI official documentation ↗Last verified: | $8 | $80 | Base |
| PremiumxAI xAI official documentation ↗Last verified: | $30 | $300 | 4× base |
| SuperxAI xAI official documentation ↗Last verified: | $100 | $1,000 | 15× base |
| UltraxAI xAI official documentation ↗Last verified: | $200 | $2,000 | 50× base |
Current self-serve plans combine Grok, Cursor and X. Paid plans add X Premium benefits when accounts are connected. Grok Bot starts at Premium. Grok models share weekly usage across supported products; third-party Cursor models use additional credits. Existing SuperGrok, Cursor and X Premium plans continue unless changed. USD list prices exclude sales tax and are separate from GlobalGPT.
xAI-Published Grok 4.7 Results
| Model / configuration | Benchmark | Published result | Model / effort conditions |
|---|---|---|---|
| Grok 4.7xAI xAI official documentation ↗Last verified: | CursorBench 4.0 | 46.3% | xhigh; Grok 4.7 release table |
| Grok 4.6xAI xAI official documentation ↗Last verified: | CursorBench 4.0 | 40.4% | high; same release comparison |
| Grok 4.7xAI xAI official documentation ↗Last verified: | Terminal-Bench 4.0 | 37.6% | xhigh; Grok 4.7 release table |
| Grok 4.6xAI xAI official documentation ↗Last verified: | Terminal-Bench 4.0 | 20.3% | high; same release comparison |
Results reported by xAI, not measured by GlobalGPT. The compared models use different reasoning efforts. Scores describe the cited evaluation harness, not every chat interaction or video model.
| Model / product | Output | Typical work | Product scope |
|---|---|---|---|
| Grok 4.7xAI xAI official documentation ↗Last verified: | Text answers, drafts and code suggestions | Demanding coding questions, source synthesis and structured planning | GlobalGPT chat; API tools require configuration |
| Grok 4.6xAI xAI official documentation ↗Last verified: | Text answers and code suggestions | Complex reasoning, source-based analysis and bounded coding | GlobalGPT chat; Grok Build is a separate product |
| Grok 4.3xAI xAI official documentation ↗Last verified: | Text answers and drafts | Previous chat option for comparing a known workflow | Previous chat model retained for comparison |
| Grok Imagine Video 1.5xAI xAI official documentation ↗Last verified: | Video | Animating a scene, a product shot or a visual concept | Dedicated GlobalGPT video workspace |
Live search and X search require tools; a model alone does not have real-time awareness. Chat, Grok Build and Imagine are distinct interfaces. The table does not claim measured response speed.
Find the Guide for Your Next Task.
Start with access or model selection, then move to a practical workflow. Each article has one primary category.
Separate subscriptions, platform access and API billing.
Give Grok a Task It Can Be Checked Against.
Choose a workflow, replace the template inputs and define what a useful answer must include.
Research & Sources
Define the source set, dates, disagreements and citation format. Use supplied sources when live search is unavailable.
Source-Based Research TemplatesWriting & Editing
Specify the audience and voice, and lock the facts and phrases that must survive the rewrite.
Writing & Editing TemplatesCoding & Debugging
Provide the runtime, failing case and expected behavior. Ask for one bounded change and run independent tests.
Coding Prompt TemplatesImage & Video Direction
Describe the subject, composition, movement and sound. A text prompt or storyboard is not a generated image or video.
Media Prompt TemplatesCreate Videos with Grok Imagine.
Start with the Imagine generator, animate a still image or create short clips for YouTube and TikTok. The task-specific tools also offer other video models.
01Grok Imagine Video Generator
Create a video from a scene idea or reference image, then review the motion and result.
GlobalGPT
02AI Image Animator
Turn a still image into a moving scene with Grok Imagine or another available video model.
GlobalGPT
03AI YouTube Video Generator
Create short video assets for a YouTube project with Grok Imagine and other available models.
GlobalGPT
04AI TikTok Video Generator
Generate short video assets for TikTok with Grok Imagine or another available model.
GlobalGPT
Check the Model, the Tools and the Evidence.
Official documentation describes model and API capabilities. GlobalGPT guides describe platform workflows and the scope of their reported tests.
Official Documentation
Grok 4.7 Developer GuideOfficial model positioning and API details; GlobalGPT chat has a separate tool surface.Grok Model CatalogCurrent model families and API options.Grok API Release NotesDated model and tool changes.Web Search ToolLive retrieval is a tool capability, not guaranteed by the model name.X Search ToolOfficial API documentation for searching X content.The X Subscription PricingOfficial provider · 2026-10-28Grok 4.7 Launch and EvaluationOfficial provider · 2026-10-28xAI API PricingOfficial provider · 2026-10-28Grok Questions, Answered
Model choice, access routes and evidence in one place.
Which Grok model should I use first?
Start with Grok 4.7 for demanding coding and knowledge work. Grok 4.6 remains available for comparison; use the same source material and acceptance criteria.
Does Grok automatically know what is happening now?
No. xAI documents that real-time information depends on search tools. Use sources and dates, and distinguish the tools available in GlobalGPT from the native Grok app or developer API.
Is Grok Imagine Video 1.5 a chat model?
No. It is a video-generation route with its own workspace and controls. The Grok chat cards are for text answers and code suggestions.
Does GlobalGPT chat include Grok Build?
Grok Build is a separate coding product. A GlobalGPT Grok chat entry does not grant the native product or prove access to its coding-agent tools.
Are xAI API prices part of a GlobalGPT subscription?
The API table quotes direct xAI developer token rates. GlobalGPT usage, native Grok subscriptions and API billing have separate account rules.
Do these benchmark scores guarantee my result?
No. xAI reported the scores under specific reasoning efforts and test conditions. Test the model on your own task and review important claims and generated code.
Bring Your Next Brief, Source Pack or Coding Question.
Start with Grok 4.7, give it a clear task and verify the result. Compare with 4.6 or switch to Imagine when the output should be video.