BREAKING
xAI Launches Grok 4.5 for Coding
0
K
context tokens
0
tok/s
generation speed
0
x
Cursor usage week 1
Coding Benchmarks
Terminal-Bench 2.1
83.3
SWE-Bench Pro
64.7
Coding Agent Index
76
0
$
input / 1M
0
$
output / 1M
0
$
cached input / 1M
Strengths and Cautions
Praised
●
Handles large codebases
●
Fewer tokens in agent loops
●
Cheaper and faster than Opus
Cautions
●
Higher cost above 200K
●
EU access mid-month
●
Post-trial pricing unclear
Check Latest Official Info
AI NEWS BLITZ
xAI has released Grok 4.5, a model tuned for coding and agentic work.