Introducing GPT-5.5 with Box

YouTube AI Channels Models

Summary

GPT-5.5 brings a 19 percentage point improvement in multi-step reasoning and financial modeling, significantly reducing the burden of knowledge work, which excites the Box team.

No content available
Original Article
View Cached Full Text

Cached at: 05/08/26, 06:28 AM

TL;DR: GPT-5.5 delivers a 19 percentage point improvement in multi-step reasoning and financial modeling, significantly reducing the burden of knowledge work, and exciting the Box team. ## The shock of the evaluation results When I saw GPT-5.5’s evaluation results come back, I couldn’t help but double-check: “Is this real?” Because compared to what I’ve seen before, this is a massive leap. I verified it several times before I was convinced of the data’s authenticity. ## A game changer for finance GPT-5.5 is an absolute game changer for our customers, especially those in the financial sector. It performs exceptionally well in multi-step reasoning. In a complex knowledge use case involving finance, GPT-5.5 stood out — it can actually make financial model predictions based on both structured and unstructured data. Compared to previous versions, this represents a **19 percentage point improvement**. ## A new level for knowledge work These improvements we’re seeing in GPT-5.5 are critical because they essentially reduce the burden of knowledge work, enabling knowledge workers to be more efficient and deliver results with higher quality and accuracy. I am incredibly excited about what GPT-5.5 will bring to Box Agent and to our customers. --- Source: Introducing GPT-5.5 with Box (https://www.youtube.com/watch?v=9cenXVrQZNQ)

Similar Articles

Introducing GPT-5.5 with Databricks

YouTube AI Channels

OpenAI partners with Databricks to release the GPT-5.5 model, achieving a 46% reduction in error rate in agent frameworks, becoming the only model to exceed 50% on benchmarks, with significant improvements in parsing quality and function calling capabilities.

Introducing GPT-5.2

OpenAI Blog

OpenAI introduces GPT-5.2, the most capable model series yet, with significant improvements in knowledge work, code generation, image perception, long-context understanding, and tool-calling. The GPT-5.2 Thinking variant achieves state-of-the-art performance on professional benchmarks, outperforming human experts on 70.9% of GDPval tasks across 44 occupations.

Introducing GPT-5.1 for developers

OpenAI Blog

OpenAI releases GPT-5.1, a new model in the GPT-5 series that dynamically adapts thinking time based on task complexity, offering 2-3x faster performance than GPT-5 while maintaining frontier intelligence. The release includes extended prompt caching (24-hour retention), new coding tools (apply_patch and shell), and a 'no reasoning' mode for latency-sensitive applications.

Introducing GPT-5.5

OpenAI Blog

OpenAI has released GPT-5.5, a significant upgrade to its frontier AI model, boasting superior capabilities in agentic coding, research, and multi-step task execution while maintaining efficiency and speed.