AI Benchmark Model
GPT-5.2 (xhigh)
Benchmark scores, pricing, speed, and model comparisons for GPT-5.2 (xhigh).
Key scores
GPT-5.2 (xhigh) is an artificial intelligence model developed by OpenAI and released on December 11, 2025. The model demonstrates specific performance metrics across various domains, achieving an intelligence index of 51.3, a coding index of 48.7, and a math index of 99. Benchmark results include an MMLU Pro score of 0.874, a GPQA score of 0.903, and an AIME 2025 score of 0.99. Regarding operational performance, the model maintains an output speed of 68.412 tokens per second with a time to first token of 68.593 seconds. Pricing for the model is structured at $1.75 per million tokens for input and $14.00 per million tokens for output, resulting in a blended rate of $4.81 per million tokens.
Benchmark results
10 tracked benchmark rows for GPT-5.2 (xhigh).
AIME 2025 Math
99.0% GPQA Reasoning
90.3% LiveCodeBench Code
88.9% MMLU Pro Knowledge
87.4% TAU2 Other
84.8% IFBench Other
75.4% LCR Other
72.7% SciCode Other
52.1% TerminalBench Hard Other
47.0% HLE Other
35.4%
Comments (0)
to join the discussion
No comments yet
Be the first to share your thoughts!