Grok 4.3: What to know
x-ai/grok-4.3Grok 4.3 is a SpaceXAI reasoning model that takes text, images, and files and returns text.
1M tokens of context — about 750,000 words, on a scale that ends at 2M.
What it can do
- Reasoning
- Vision
- Tool use
- Takes text
- Takes image
- Takes file
- Gives text
Released 30 Apr 2026 · checked 3 Oct 2026.
About it
Grok 4.3 is a reasoning model from SpaceXAI. It can take text, images, and files as input and produces text. Its listed strengths include agent-style workflows, following instructions, and tasks that need factual answers. It has a 1-million-token context window and supports tools. Input and output are priced separately per million tokens.
Review
In a line
Grok 4.3 is a reasoning model made by SpaceXAI. It accepts text, images, and files, and responds with text.
What it's good at
The model is suited to workflows where it uses tools, follows instructions, or handles tasks that need factual answers. It supports image input as well as text and files.
It may suit readers comparing reasoning models or models that work with images.
Price in plain terms
Input costs $1.25 per million tokens, and output costs $2.50 per million tokens. A token is a small piece of text processed by the model.
At the stated rates, reading a roughly 500-word page costs about 0.081 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.25 cents; 1,000 such turns cost about $2.50.
Context and limits
Grok 4.3 can take in up to 1 million tokens at once and produce up to 900,000 tokens in one answer.
It returns text. The listed inputs are text, images, and files; the facts do not specify other output types.
Ways to use it
The model supports tools and accepts text, images, and files. Its model ID is x-ai/grok-4.3.
Verdict
Consider Grok 4.3 if you need a reasoning model for tool-using workflows, instruction-following, or factual-answer tasks, and want to provide images or files as well as text. Its large context window may suit work that requires taking in a lot of material at once. Compare its separate input and output charges with your expected use. If you need an answer format other than text, this model's listed output is text.
Verdict
Grok 4.3 is worth considering for people who need a model to reason through instructions, use tools, or work with image and file inputs. Its 1-million-token context and high output limit make it suited to tasks involving substantial material. Costs are low for the example chat turns listed, though actual spending depends on how many tokens go in and come out. Skip it if you need non-text answers or require confirmed access through a particular app; availability details are not provided here.
- Accepts text, images, and files as input
- Supports tool use and reasoning
- Has a 1-million-token context window
- Input and output rates are listed separately
- Produces text only
- Actual cost depends on input and output token use
- Pick it ifTool-using workflows and instruction-following tasks; Tasks that need factual answers or image input; Readers comparing reasoning models
- Skip it ifYou need output in a format other than text; You need confirmed availability in a particular app
Pricing
Input is $1.25 per million tokens and output is $2.50 per million tokens. The example cost is about 0.25 cents for one chat turn, or about $2.50 for 1,000 such turns.
Questions
Who makes Grok 4.3?
Grok 4.3 is made by SpaceXAI.
What can I give Grok 4.3 as input?
It accepts text, images, and files.
What kind of output does it produce?
Its listed output is text.
How much does Grok 4.3 cost?
Input costs $1.25 per million tokens and output costs $2.50 per million tokens.
How large is its context window?
It can take in up to 1 million tokens at once.
Can Grok 4.3 use tools?
Yes. Tool support is listed for the model.