ColinBuilds.com · Official xAI API pricing
Grok 4.3
- Input / 1M
- $1.25
- Output / 1M
- $2.50
- Official xAI context window
- 1,000,000 tokens
xAI grok-4.3 official API pricing; input $1.25/1M tokens, cached input $0.20/1M tokens, output $2.50/1M tokens. xAI notes higher context pricing for requests exceeding 200K tokens. · as of July 4, 2026
Compare models
Same per-1M input/output rates as above. Enter your monthly token volumes to compare bills.
Grok 4.3
- Input: 1M × $1.25 = $1.25
- Output: 1M × $2.50 = $2.50
- Total
- $3.75
GPT-4o
- Input: 1M × $2.50 = $2.50
- Output: 1M × $10.00 = $10.00
- Total
- $12.50
Difference
$8.75
GPT-4o costs $8.75 more than Grok 4.3.
Popular comparisons
30 total featuring Grok 4.3.
Grok 4.3 is xAI’s current official Grok API model name for the Grok 4 line in the xAI developer docs, with text and image input, text output, a 1M token context window, and official xAI token pricing.
What this model is
xAI lists the exact model name as grok-4.3, with aliases grok-4.3-latest and grok-latest. The model page describes Grok 4.3 as xAI’s advanced flagship model for non-hallucination rate, agentic tool calling, and instruction following.
For builders, the important detail is the exact API identity. A request for “Grok 4” should be checked against xAI’s current docs before writing pricing or model IDs, because the official model page currently exposes grok-4.3 rather than a usable grok-4 page.
Pricing notes
xAI’s official Grok 4.3 model page lists API pricing at $1.25 per 1M input tokens, $0.20 per 1M cached input tokens, and $2.50 per 1M output tokens.
The calculator on this page uses the base uncached input and output rates: $1.25 input and $2.50 output per 1M tokens. xAI also notes that higher context pricing applies for requests exceeding 200K tokens, so very long-context production estimates should check the xAI pricing page directly before budgeting.
Benchmarks and specs
xAI’s model page lists Grok 4.3 with text and image input, text output, a 1,000,000 token context window, function calling, structured outputs, and configurable reasoning levels of none, low, medium, and high.
This page does not show a parameter count or public benchmark score. Those numbers should only be shown when exact public sources are matched to grok-4.3 specifically.
Best fit
Grok 4.3 is best for builders who want xAI’s official flagship API route for assistants, tool-using agents, structured-output workflows, image-plus-text prompts, and long-context tasks.
It is also a useful cost comparison against other official frontier APIs because its base listed input and output rates are easy to compare per 1M tokens, while cached input and higher-context usage can change the real bill.