Box
We tested @OpenAI's new GPT-5.4 model, and it showed a 78% overall extraction accuracy - up 6 points from GPT-5.2. Our evaluation tested the model across real industries and document workflows: → Clinical data: +5 pts (81% → 86%) → Legal agreements: +3 pts (82% → 85%) → Regulatory filings: +5 pts (79% → 84%) → Research publications: +7 pts (71% → 78%) → Government statistical Publications: +10 pts (60% → 70%) → Industry & media reports: +7 pts (61% → 68%) GPT-5.4 excels at handling complex reasoning, dense reports, and inconsistent document structures with reliability. Dive into our full evaluation → https://bit.ly/4cvqSnE GPT-5.4 will be available in Box AI Studio today.
