AI Benchmark Model

GPT-5.4 (xhigh)

Benchmark scores, pricing, speed, and model comparisons for GPT-5.4 (xhigh).

Key scores

Intelligence56.8Index score
Coding57.2Index score
Speed203.2Tokens/sec
Blended Cost$5.63Per 1M tokens

GPT-5.4 (xhigh) is an artificial intelligence model developed by OpenAI and released on March 5, 2026. In performance evaluations, the model achieved an intelligence index of 56.8 and a coding index of 57.2. Benchmark results include a GPQA score of 0.92, an HLE score of 0.416, a SciCode score of 0.566, and an IFBench score of 0.739. Additionally, the model recorded a LCR score of 0.74, a TerminalBench Hard score of 0.576, and a TAU2 score of 0.871. Pricing for this model is set at $2.50 per 1 million input tokens and $15.00 per 1 million output tokens, resulting in a blended rate of $5.63. Operational performance metrics indicate an output speed of 78.88 tokens per second.

Benchmark results

7 tracked benchmark rows for GPT-5.4 (xhigh).

GPQA Reasoning
92.0%
TAU2 Other
87.1%
LCR Other
74.0%
IFBench Other
73.9%
TerminalBench Hard Other
57.6%
SciCode Other
56.6%
HLE Other
41.6%

Comments (0)

No comments yet

Be the first to share your thoughts!