Creator

OpenAI

Released

2024-12-05

Intelligence

23.4

Artificial Analysis Index

Coding

39.7

Artificial Analysis Index

In $/1M

$15.00

input tokens

Out $/1M

$60.00

output tokens

Blended $/1M

$26.25

3:1 blended

Speed

0

tokens / sec

Profile

License, openness, and modality — the governance layer.

License

Proprietary

non-commercial

Weights

Closed

API only

Modalities

Text

Parameters

A research preview model focused on mathematical and logical reasoning capabilities, demonstrating improved performance on tasks requiring step-by-step reasoning, mathematical problem-solving, and code generation. The model shows enhanced capabilities in formal reasoning while maintaining strong general capabilities.

Capability profile

Category strength across 20 domains, via LLM Stats.

legal
90finance
90language
90math
80biology
80physics
80chemistry
80healthcare
80vision
70
general
70reasoning
70multimodal
70code
60tool calling
60communication
60factuality
50frontend development
40
Via LLM Stats

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
75
Humanity’s Last Exam
8
MMLU-Pro
84
SciCode
36
LiveCodeBench
68
MATH-500
97
AIME 2025
τ²-Bench (agentic)
63
Terminal-Bench Hard
13
IFBench
70
Via Artificial Analysis
© 2026 NYSGPT2525 LLC