GPT-5.4 mini

OpenAI

OpenAI's strongest mini model yet (April 2026) for coding, computer use, and subagents. A low-cost, high-throughput option with a 400K-token context window and multimodal (text + vision) input.

Strengths

Strongest OpenAI mini model for coding, computer use, and subagents
Very low cost ($0.75/$4.50) for high-volume workloads
400K-token context window with multimodal (text + vision) input

Caveats

Lower capability ceiling than GPT-5.4 / 5.5 for complex tasks
Best as a subagent / high-throughput model rather than a primary reasoner

Vendor: OpenAI
Context Window: 400,000 tokens
Max Output: 128,000 tokens
License: Proprietary
Release Date: 2026-04-23
Training Cutoff: 2025-08-31
Modalities: text
image

Capabilities

Vision

Audio

Video

Tool Use

Resources

API Documentation

Pricing

$0.75 per 1M input tokens, $4.50 per 1M output tokens - low cost for high-volume coding and subagent workloads.