xAI
Coding and knowledge-work model with 500K context, image input, and configurable reasoning effort.
At a glance
API model ID: grok-4.7
Capabilities & boundaries
Tool support requires the appropriate API integration; a supported tool is not automatically active in every chat.
Reasoning cannot be disabled. Start with low for focused steps, then evaluate medium or high for harder work. xhigh trades more latency for additional reasoning; compare results on your own examples rather than assuming the highest setting is always necessary.
Unsupported settings: none.
Put it to work
Original examples from EZ Ai Assist. Adapt these to your task and the features available in your workspace.
Workflow 01
Keep a long change grounded in the actual code and acceptance criteria.
Using the repository files and migration requirements I provide, map the affected modules and propose the smallest staged migration. For each stage, list the code changes, compatibility risks, tests and rollback condition. Cite file paths for your findings. Separate confirmed dependencies from assumptions and do not claim any test was run unless you actually ran it.
Workflow 02
Turn supplied business material into a decision others can audit.
Create a decision brief from the documents below. Define the decision, compare the options against the stated constraints, and cite the document and section behind each important claim. Include a recommendation, the strongest counterargument, unresolved evidence gaps and a short validation plan. Do not substitute outside facts for missing evidence.
Workflow 03
Use visual input to produce actionable acceptance criteria.
Inspect this product screenshot and the requirements I include. Describe the visible user journey, identify mismatches with the requirements, and turn each finding into a testable acceptance criterion. Separate visible evidence from assumptions about interactions. Prioritize the three changes with the clearest user benefit and explain how I can verify them.
Developer reference
These are global xAI API reference prices, not EZ Ai Assist subscription prices.
View EZ Ai Assist plans| Token type | < 200,000prompt tokens | ≥ 200,000prompt tokens |
|---|---|---|
| Input | $2.00 | $4.00 |
| Cached input | $0.50 | $1.00 |
| Output | $6.00 | $12.00 |
| Usage | Rate and unit |
|---|---|
| Web Search | $5.00per 1,000 calls |
| X Search: posts | $5.00per 1,000 posts fetched |
| X Search: profiles | $10.00per 1,000 profiles fetched |
| Code execution | $5.00per 1,000 calls |
Common questions
xAI positions it for coding, agentic tasks and knowledge work. A practical evaluation is a multi-file change or a document-backed brief with explicit tests and acceptance criteria. Its usefulness still depends on the evidence and tools supplied.
The public card lists 500,000 tokens of context. That is not a 500,000-token output allowance. The reviewed sources do not verify a separate output ceiling or training cutoff, and an app integration may impose smaller limits.
No. The documented efforts are low, medium, high and xhigh, with high as the default. Evaluate the latency and usage difference before selecting xhigh for a recurring workflow.
It accepts text and images and returns text. That supports visual analysis, not a promise of native image, audio or video generation. Image upload support in EZ Ai Assist must be checked separately.
No. Search and other server-side tools need a supported integration that enables them. Tool output can improve freshness, but sources and claims still need review. Tool calls or fetched items also add charges.
At 200,000 prompt tokens, the entire request moves to the higher input, cached-input and output rates. Reasoning and enabled tools can increase the total beyond the visible answer alone.
This page describes xAI API reference behavior, not EZ Ai Assist billing. Fast is a Cursor/Grok Build offering, not a separate public API option. Check EZ Ai Assist for the actual model access, controls and subscription terms available to you.
Check the source
Specifications and API prices checked on . Example prompts and workflow advice are editorial guidance from EZ Ai Assist.
Same provider
xAI
Coding and agentic model with 500K context, structured outputs, and four reasoning effort levels.
xAI
Coding and engineering model with 500K context, image input, tool calling, and configurable reasoning.
xAI
1M-context model for instruction-driven work and tool calling, with configurable reasoning that can be disabled.
xAI
Coding-focused model with 256K context, image input, function calling, and structured outputs.