GPT-5.6 Terra

Routed in Core
by OpenAI

Strong daily coding and knowledge work at a moderate cost.

Intelligence55Artificial Analysis Index
Average task cost$0.82Observed benchmark task
Cost per million$2.5 in$15 out
Output speed140tokens per second
Evidence

What we know

Open source
Availability
Available through Core's model router
Core fit
52 / 100 · Efficient
Public intelligence
55 · max effort
Benchmark snapshot
Jul 2026
Input price
$2.5 per 1M tokens
Output price
$15 per 1M tokens
Cache read
$0.25 per 1M tokens