Grok 4.3
Grok 4.3 combines a one-million-token context with text and image input. It supports a no-reasoning setting as well as reasoning effort levels.
What the release provides
Reasoning effort can be none, low, medium, high or xhigh; low is the default. Function calling, structured output and batch processing are listed.Grok 4.3 API model card
How to use it well
Practical guidance. For document extraction, begin with a clear schema and a small relevant passage. Increasing the input to the full context limit can hide the evidence the answer needs. Use the no-reasoning setting only after checking the required fields and edge cases.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | grok-4.3 |
| Context window | 1,000,000 tokens Published API context limit Grok 4.3 API model card |
| Input price | 1.25 USD per million tokens Standard uncached input; prompt below 200,000 tokens; higher rate applies to all tokens at the threshold Grok 4.3 API model card |
| Output price | 2.5 USD per million tokens Standard output; prompt below 200,000 tokens; excludes tools and batch discounts Grok 4.3 API model card |