OpenAI Responses API Now Accepts Word, Excel, and PowerPoint Files

OpenAI

Feb 24, 2026 · Updated Apr 25, 2026

OpenAI's Responses API now accepts Word, PowerPoint, Excel, CSV, and other office files directly as inputs. Agents can pull context from real business documents without conversion pipelines, enabling tasks like analyzing reports or extracting data from spreadsheets mid-workflow.

OpenAI expanded file input support in the Responses API to include .docx, .pptx, .xlsx, .csv, .tsv, .txt, and code files - formats agents encounter constantly in real-world enterprise workflows. PDF support already existed; this closes the gap for the office document formats that make up the bulk of business data.

The API handles each type differently: text is extracted from Word and PowerPoint files, while spreadsheets get a special augmentation flow optimized for structured data. One limitation: embedded charts and images in non-PDF files aren't extracted - convert to PDF first if you need visual fidelity.

This significantly reduces pre-processing overhead for document-heavy agents. Pass a .docx meeting summary or an .xlsx financial report directly in your API call and let the model pull context without an intermediate extraction step.

View the full update on developers.openai.com

OpenAI Developers

@OpenAIDevsFeb 24

We expanded file input types so you can now pass docx, pptx, csv, xlsx, and more directly to the Responses API. Your agents can now pull context from real-world files and generate more accurate outputs. https://t.co/brKUsSfOJy https://t.co/no5zPgWgSv

118

View on X

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from OpenAI →

Keep reading

OpenAI Equips Responses API with a Full Computer Environment

OpenAI's Responses API now runs agents inside hosted containers with file systems, network access, and a shell tool. The agent loop handles long-running workflows by proposing shell commands, streaming results back, and compacting context automatically when the window fills.

LangChainJun 9

LangSmith Fleet Adds Direct File Interaction for Agent-Driven Content Creation

LangChain's LangSmith Fleet now supports direct file interaction, allowing AI agents to create and edit documents, presentations, and webpages. This update enables agents to generate new content or collaboratively edit existing files directly within conversations, streamlining content workflows.

Google Gemini Generates Downloadable Office Files Directly From Your Chat Prompts

GeminiApr 30

Google Gemini Generates Downloadable Office Files Directly From Your Chat Prompts

Google updated the Gemini app to generate downloadable files, including PDFs, Google Workspace documents, and Microsoft Office files, directly from conversational prompts. This removes the need to copy-paste AI responses into external editors or upload templates to get a finished work product. By producing native file formats like .docx and .xlsx, Gemini shifts from a text assistant to a functional tool that delivers ready-to-share documents.

PerplexityMay 28

Perplexity Integrates Multi-Model Computer Agent Into Core Microsoft 365 Apps

Perplexity launched native integrations for Word, Excel, PowerPoint, and Outlook, embedding its agentic orchestration layer directly into the Microsoft 365 side panel. The update eliminates manual copy-pasting by allowing users to generate research-backed documents, financial models, and slide decks where they already work.