LLM

GPT-5.4 mini (xhigh)

Name: GPT-5.4 mini (xhigh) Review & Benchmarks
Item: GPT-5.4 mini (xhigh)
Rating: 24.1
Author: Sentient Weekly

by OpenAI

Overall Score

24.1

Released Mar 2026

Benchmark Scores

Reasoning

—

Coding

56.1

Math

—

Creative writing

—

Instruction following

—

Multimodal

—

Standard Benchmarks

aa_gpqaaa_gpqa

0.9

Measured 2026-09-26 · source

aa_hleaa_hle

0.3

Measured 2026-09-26 · source

aa_scicodeaa_scicode

0.5

Measured 2026-09-26 · source

aa_ifbenchaa_ifbench

0.7

Measured 2026-09-26 · source

aa_lcraa_lcr

0.8

Measured 2026-09-26 · source

aa_tau2aa_tau2

0.8

Measured 2026-09-26 · source

aa_tau_bankingaa_tau_banking

0.3

Measured 2026-09-26 · source

aa_terminalbench_hardaa_terminalbench_hard

0.5

Measured 2026-09-26 · source

aa_terminalbench_v2_1aa_terminalbench_v2_1

0.6

Measured 2026-09-26 · source

Price

—

Context

—

Speed

183.8 t/s

2613ms TTFT

Compare this model →

Discussion

0 comments

Be the first to comment.