ModelsCompareBest forBenchmarksStatusPricingAPI

GPT-5.4 nano

OpenAI·Operational·Released 2026-03-17textvision
Context400k
Max output128k
Input / 1M$0.2
Output / 1M$1.25
Cached in$0.02
Benchmarks10

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

GPQA Diamond#40/7082.8%
OSWorld-Verified#12/1239%
SWE-Bench Pro#14/1552.4%
Terminal-Bench 2.0#11/1346.3%
Toolathlon#6/635.5%

Access GPT-5.4 nano and every other model through one endpoint with automatic failover: Respan gateway.