← Model indexMODEL PROFILE / grok-4-3

xAI / current

Grok 4.3

xAI's recommended general chat model for reasoning, coding and tool-augmented current-information workflows.

Context1M
ModalityText · Image
AccessAPI
ReasoningConfigurable

Where it fits.

Best for time-sensitive information tasks when paired with xAI search tools; realtime knowledge is not automatic. Grok 4.3 combines text and image input with a 1M context, function calling, structured outputs and configurable reasoning from none to high. Realtime knowledge comes from explicitly enabled Web or X search.

02

Core capabilities

  • 1M context for text and image input
  • Reasoning settings: none, low, medium and high
  • Function calling and structured outputs
  • Optional server-side Web Search and X Search
03

Best-fit workflows

  • Search-augmented Q&A
  • General analysis and coding
  • Realtime social signals
04 / OPERATIONS

From evaluation
to production.

Deployment notes

Use the exact grok-4.3 ID when stability matters and treat latest aliases as rolling targets. Keep search opt-in by workflow, validate citations, cache reusable context and test cost at both sides of the 200K threshold.

Watch before choosing

Realtime behavior requires explicit Web/X search tools, along with source-quality and citation controls. Requests beyond 200K context use a higher pricing regime. Search-enabled answers need source-quality controls, citation display and defenses against untrusted retrieved content.

05

Family & versions

Specifications and positioning checked against provider documentation.

Open official source↗