OpenAI

gpt-5.2-2025-12-11

proprietaryMultimodal

Benchmark Average (0–100)

77.4%

Average of non-null benchmark scores across all evaluated tasks.

BENCHMARK SCORES

GPQA92.4%

Graduate-Level Google-Proof Q&A. PhD-level multiple-choice questions in chemistry, biology, and physics. Scored by accuracy.

AIME 2025100.0%

AIME 2025 is a mathematics competition benchmark testing advanced problem-solving.

SWE-bench Verified80.0%

SWE-bench Verified evaluates models on real-world software engineering tasks from GitHub issues.

MMMLU89.6%

Multilingual MMLU tests knowledge across many languages and subject areas.

BrowseComp65.8%

BrowseComp tests web browsing comprehension, measuring ability to find and synthesize information from websites.

HLE34.5%

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-Pro79.5%

MMMU-Pro is a more challenging version of MMMU with harder multimodal understanding tasks.

OTHER BENCHMARKS

ToolAthalon

46.3%

ARC-AGI v2

52.9%

CharXIV-R

82.1%

ScreenSpot Pro

86.3%

MCP Atlas

60.6%

FrontierMath

40.3%

MODEL INFO

Organization

OpenAI

Context Window

400,000

Release Date

2025-12-11

Knowledge Cutoff

2025-08-25

Parameters

—

License

proprietary

Input Price ($/M)

$1.7500

Output Price ($/M)

$14.0000

METADATA

Announcement Date

2025-12-11

Organization Country

Canonical ID

—