OpenAI: Responses API web search can now return image results
OpenAI added image results to the Responses API web_search tool, letting apps retrieve web-grounded visuals (with source links) alongside regular text results.
OpenAI says your app can now search the web for images: web search in the Responses API "now supports image results in addition to text results," enabling apps that surface "products, places, visual references, and source links for inspiration." (OpenAI Developers on X)
OpenAI’s API changelog describes this as a Responses API web search update: "Web search can now return image results alongside regular text results" and suggests using it when you need "current or web-grounded visuals, such as product photos, landmarks, places, events, or visual references." (OpenAI API changelog)
What this enables
If you already use the Responses API web_search / web_search_preview-style tooling for web-grounded text answers, this adds a new option: returning image results with attribution so you can render a visual gallery or "top images" strip next to text.
Practical use cases (all explicitly called out by OpenAI in its changelog) include:
- Product photos
- Landmarks and places
- Events
- Visual references
What we are (and aren’t) claiming
OpenAI’s announcement is about search results, not image generation.
We’re only claiming what OpenAI has stated publicly:
- The Responses API web search tool can return image results alongside text results.
- These results are intended for web-grounded visuals with source links.
We are not adding pricing, rollout, or response-schema details here because OpenAI’s developer guide link in the changelog was not accessible during this run.
Sources
- OpenAI Developers: "Your app can now search the web for images" (Jun 9, 2026) — https://x.com/OpenAIDevs/status/2064395155688616153
- OpenAI API Changelog: "Web search can now return image results" (Jun 9, 2026) — https://platform.openai.com/docs/changelog
Read next
Keep building the workspace playbookGPT-5.5 Is Here: State-of-the-Art Agentic Coding, 1M Context, and a New Pro Tier
OpenAI launched GPT-5.5 on April 23, 2026 -- its smartest model yet, with state-of-the-art scores on Terminal-Bench 2.0 (82.7%), GDPval (84.9%), and OSWorld-Verified (78.7%), GPT-5.4 per-token latency, and a new GPT-5.5 Pro tier for harder work. Available in ChatGPT and Codex today, with API at $5/M input and $30/M output coming soon.
OpenAI's Secure MCP Tunnel: Connect Private MCP Servers Over Outbound-Only HTTPS
OpenAI published a Secure MCP Tunnel guide that connects private and on-prem MCP servers to ChatGPT and Codex without opening inbound firewall ports. A tunnel-client polls OpenAI for work over outbound HTTPS, forwards JSON-RPC requests to the local server, and posts responses back through the same tunnel.
Frequently Asked Questions
What changed in OpenAI's Responses API web search?
OpenAI says the Responses API web_search tool can now return image results alongside regular text results, making it possible to build apps that surface web-grounded visuals with source links.
When should you use image results in web search?
OpenAI's API changelog recommends image results when an application needs current or web-grounded visuals like product photos, landmarks/places, events, or other visual references.