Grok 4.5 is xAI's July 2026 flagship for coding, agentic work, and knowledge tasks: 500K context, $2/1M input, $6/1M output, and $0.50 cached input.
By the benchr team · · Figures verified against official sources, July 13, 2026
Official pricing and limits
Grok 4.5 API pricing: $2 input, $6 output per 1M — official data
Field
Value
Model ID
grok-4.5
Input / 1M
$2.00
Cached input / 1M
$0.50
Output / 1M
$6.00
Context window
500,000 tokens
Knowledge cutoff
February 1, 2026
Reasoning
Configurable
Where it fits
Use Grok 4.5 when you want xAI's newest coding and agentic model with lower output pricing than the most expensive OpenAI and Anthropic flagships. It is the default xAI pick for code, chat, and knowledge work.
When to skip it
Skip it when you need a full 1M context window, where Grok 4.3 and several other models still publish larger context limits, or when procurement requires official benchmark tables that xAI has not published on the checked docs.
benchr