OpenAI
gpt-5.4-mini
proprietaryMultimodal
Benchmark Average (0–100)
64.3%
Average of non-null benchmark scores across all evaluated tasks.
BENCHMARK SCORES
GPQA88.0%
Graduate-Level Google-Proof Q&A. PhD-level multiple-choice questions in chemistry, biology, and physics. Scored by accuracy.
HLE28.2%
Humanity's Last Exam (HLE) is a multi-modal benchmark testing frontier knowledge across mathematics, humanities, and natural sciences with 2,500 expert-level questions.
MMMU-Pro76.6%
MMMU-Pro is a more challenging version of MMMU with harder multimodal understanding tasks.
OTHER BENCHMARKS
ToolAthalon
42.9%
MCP Atlas
57.7%
MODEL INFO
Organization
OpenAI
Context Window
400,000
Release Date
2026-03-17
Knowledge Cutoff
2025-08-31
Parameters
—
License
proprietary
Input Price ($/M)
$0.7500
Output Price ($/M)
$4.5000
METADATA
Announcement Date
2026-03-17
Organization Country
US
Canonical ID
—