# \#file-search **URL:** https://community.openai.com/tag/file-search/707.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [Including citations from both file\_search tool and web\_search tool in the same response](https://community.openai.com/t/including-citations-from-both-file-search-tool-and-web-search-tool-in-the-same-response/1379247)
**Author:** [@RobertoZanolli](https://community.openai.com/u/RobertoZanolli)\ **Replies:** 11\ **Last updated:** [October 11, 2026, 9:39am UTC](https://community.openai.com/t/including-citations-from-both-file-search-tool-and-web-search-tool-in-the-same-response/1379247 "2026-10-11T09:39:00Z")
Hi, I’m trying to fix an issue in a RAG system with file\_search that also uses web\_search tool restricted to a set of allowed domains (using responses api). The problem I’m facing is the following: when only one tool r… --- ## [OpenAI Platform Evals w/ gpt-5 models & file\_search tool returns "Empty assistant message"](https://community.openai.com/t/openai-platform-evals-w-gpt-5-models-file-search-tool-returns-empty-assistant-message/1360074)
**Author:** [@solosolo](https://community.openai.com/u/solosolo)\ **Replies:** 2\ **Last updated:** [September 20, 2026, 7:28pm UTC](https://community.openai.com/t/openai-platform-evals-w-gpt-5-models-file-search-tool-returns-empty-assistant-message/1360074 "2026-09-20T19:28:55Z")
I’m using the OpenAI Platform’s Evals to evaluate tool usage, specifically with the file\_search tool. When I evaluate the arguments the model passes to the tool, to make sure they are accurate, etc many of the responses … --- ## [GPT 5,6-sol - the retrieval/indexing layer is failing to expose attachment contents](https://community.openai.com/t/gpt-5-6-sol-the-retrieval-indexing-layer-is-failing-to-expose-attachment-contents/1399092)
**Author:** [@ojfd](https://community.openai.com/u/ojfd)\ **Replies:** 1\ **Last updated:** [September 19, 2026, 4:28pm UTC](https://community.openai.com/t/gpt-5-6-sol-the-retrieval-indexing-layer-is-failing-to-expose-attachment-contents/1399092 "2026-09-19T16:28:45Z")
Environment Surface: Web Browser version: Firefox Affected feature and model, if relevant: GPT 5,6-sol Plus subscription Assistant is unable to retrieve the attachment contents. These are two assistant responses. “… --- ## [Responses API: strict json\_schema returns malformed JSON when combined with file\_search + \`include: file\_search\_call.results\`](https://community.openai.com/t/responses-api-strict-json-schema-returns-malformed-json-when-combined-with-file-search-include-file-search-call-results/1380608)
**Author:** [@stevevisokio](https://community.openai.com/u/stevevisokio)\ **Replies:** 1\ **Last updated:** [September 17, 2026, 8:50pm UTC](https://community.openai.com/t/responses-api-strict-json-schema-returns-malformed-json-when-combined-with-file-search-include-file-search-call-results/1380608 "2026-09-17T20:50:01Z")
Summary With text.format set to json\_schema + strict: true, and tools including file\_search (a vector store) and web\_search, and include containing file\_search\_call.results, the Responses API intermittently returns mal… --- ## [Feature Request: Managed Hot Context / Agent Memory Store](https://community.openai.com/t/feature-request-managed-hot-context-agent-memory-store/1394197)
**Author:** [@isaac.holmes](https://community.openai.com/u/isaac.holmes)\ **Replies:** 1\ **Last updated:** [September 8, 2026, 9:15am UTC](https://community.openai.com/t/feature-request-managed-hot-context-agent-memory-store/1394197 "2026-09-08T09:15:26Z")
Feature Request: Managed Hot Context / Agent Memory Store I would like an OpenAI-managed data layer designed specifically for low-latency agent memory and context retrieval. When integrating AI into business application… --- ## [Feature Request: Add Project Search](https://community.openai.com/t/feature-request-add-project-search/1394326)
**Author:** [@majo.uram](https://community.openai.com/u/majo.uram)\ **Replies:** 1\ **Last updated:** [September 4, 2026, 1:02pm UTC](https://community.openai.com/t/feature-request-add-project-search/1394326 "2026-09-04T13:02:26Z")
Please add a search bar to the Projects list. When users have many projects, finding the correct one becomes slow and increases the risk of opening the wrong project. A simple search by project name or partial name wou… --- ## [PDF file uploaded to vector store cannot be used by the model](https://community.openai.com/t/pdf-file-uploaded-to-vector-store-cannot-be-used-by-the-model/1392435)
**Author:** [@DanTheManIsBack](https://community.openai.com/u/DanTheManIsBack)\ **Replies:** 7\ **Last updated:** [August 30, 2026, 12:22pm UTC](https://community.openai.com/t/pdf-file-uploaded-to-vector-store-cannot-be-used-by-the-model/1392435 "2026-08-30T12:22:24Z")
I have 5 files uploaded to a vector store, all of which have the status completed. However when prompted to return all file names, it misses out one. Then after some debugging, I found out that the file has the usage\_byt… --- ## [API Returns 424 Error for MCP Calls](https://community.openai.com/t/api-returns-424-error-for-mcp-calls/1353593)
**Author:** [@guytp](https://community.openai.com/u/guytp)\ **Replies:** 5\ **Last updated:** [August 24, 2026, 2:43pm UTC](https://community.openai.com/t/api-returns-424-error-for-mcp-calls/1353593 "2026-08-24T14:43:49Z")
I’m having an issue with integrating an MCP server during testing. To try and simplify this I am at the point of the simplest schema and request for it and I’ve removed all our code and am just testing through the API p… --- ## [Extremely inconsistent vector store indexing times](https://community.openai.com/t/extremely-inconsistent-vector-store-indexing-times/1388135)
**Author:** [@Neil\_Lewis](https://community.openai.com/u/Neil_Lewis)\ **Replies:** 1\ **Last updated:** [July 29, 2026, 7:09pm UTC](https://community.openai.com/t/extremely-inconsistent-vector-store-indexing-times/1388135 "2026-07-29T19:09:46Z")
@OpenAI\_Support I am seeing extreme, intermittent latency in OpenAI-managed vector-store ingestion. I ran the same capability test twice using the same API project, credentials, endpoint, and ingestion configuration. E… --- ## [Responses API file\_search intermittently returns zero results for the same prompt](https://community.openai.com/t/responses-api-file-search-intermittently-returns-zero-results-for-the-same-prompt/1380262)
**Author:** [@makro](https://community.openai.com/u/makro)\ **Replies:** 4\ **Last updated:** [July 13, 2026, 2:06pm UTC](https://community.openai.com/t/responses-api-file-search-intermittently-returns-zero-results-for-the-same-prompt/1380262 "2026-07-13T14:06:12Z")
I’m building a RAG assistant using the Responses API with the file\_search tool and an existing vector store. I’m seeing intermittent false-negative retrieval. The same prompt, sent in separate new chats, sometimes retri… --- ## [File Citations bug in file\_search citations using gpt-5.5](https://community.openai.com/t/file-citations-bug-in-file-search-citations-using-gpt-5-5/1383150)
**Author:** [@tristan.mathieson](https://community.openai.com/u/tristan.mathieson)\ **Replies:** 13\ **Last updated:** [June 14, 2026, 2:12pm UTC](https://community.openai.com/t/file-citations-bug-in-file-search-citations-using-gpt-5-5/1383150 "2026-06-14T14:12:32Z")
For the past several days, the our documentation chatbot (powered by gpt-5.5 using the file\_search tool) has not been correctly adding the citations in its responses. For example, here is a recent output message: {‘id’:… --- ## [Bug: Vector store status: completed does not guarantee searchability - file\_search returns empty results silently](https://community.openai.com/t/bug-vector-store-status-completed-does-not-guarantee-searchability-file-search-returns-empty-results-silently/1374471)
**Author:** [@morritse](https://community.openai.com/u/morritse)\ **Replies:** 6\ **Last updated:** [May 12, 2026, 9:15pm UTC](https://community.openai.com/t/bug-vector-store-status-completed-does-not-guarantee-searchability-file-search-returns-empty-results-silently/1374471 "2026-05-12T21:15:36Z")
Bug: Vector store status: completed does not guarantee searchability - file\_search returns empty results silently Summary After creating a vector store with vectorStores.create({ file\_ids: \[…\] }) and polling until status… --- ## [Uploading large numbers of files and searching - whats the best way](https://community.openai.com/t/uploading-large-numbers-of-files-and-searching-whats-the-best-way/1376448)
**Author:** [@Sunny996](https://community.openai.com/u/Sunny996)\ **Replies:** 1\ **Last updated:** [March 12, 2026, 9:51am UTC](https://community.openai.com/t/uploading-large-numbers-of-files-and-searching-whats-the-best-way/1376448 "2026-03-12T09:51:05Z")
I want to upload and search on a large number of files whats the best way I , am getting a max file limit --- ## [Vector Store file ingestion failing consistently for .docx files (server\_error)](https://community.openai.com/t/vector-store-file-ingestion-failing-consistently-for-docx-files-server-error/1372318)
**Author:** [@Lorenz1](https://community.openai.com/u/Lorenz1)\ **Replies:** 1\ **Last updated:** [January 23, 2026, 5:18pm UTC](https://community.openai.com/t/vector-store-file-ingestion-failing-consistently-for-docx-files-server-error/1372318 "2026-01-23T17:18:01Z")
Hi everyone, I am encountering a persistent issue when trying to add .docx files to a Vector Store. I have tested this via both the API and the OpenAI Platform UI (dashboard), and the result is the same: .docx files fa… --- ## [Responses API file\_search: later tool-call ranking/score seems biased by earlier topic even with “one-topic-per-call”](https://community.openai.com/t/responses-api-file-search-later-tool-call-ranking-score-seems-biased-by-earlier-topic-even-with-one-topic-per-call/1371800)
**Author:** [@s-nishihara1](https://community.openai.com/u/s-nishihara1)\ **Replies:** 0\ **Last updated:** [January 16, 2026, 10:27am UTC](https://community.openai.com/t/responses-api-file-search-later-tool-call-ranking-score-seems-biased-by-earlier-topic-even-with-one-topic-per-call/1371800 "2026-01-16T10:27:43Z")
I’m building a hotel guest-support (concierge) system using the Responses API with a Vector Store + built-in file\_search tool. What I’m trying to do Guests often ask multiple unrelated questions in a single message (e.… --- ## [Inconsistent unsupported\_file error when adding files to Vector Store](https://community.openai.com/t/inconsistent-unsupported-file-error-when-adding-files-to-vector-store/1371682)
**Author:** [@Asad\_Ullah1](https://community.openai.com/u/Asad_Ullah1)\ **Replies:** 0\ **Last updated:** [January 14, 2026, 5:03pm UTC](https://community.openai.com/t/inconsistent-unsupported-file-error-when-adding-files-to-vector-store/1371682 "2026-01-14T17:03:10Z")
I’m seeing an inconsistent issue when adding a file to a Vector Store. The same file sometimes fails with the error: unsupported\_file: The file type is not supported and at other times works fine on the same day. This… --- ## [Connecting to GPTs through API (again)](https://community.openai.com/t/connecting-to-gpts-through-api-again/1367761)
**Author:** [@WillEndure](https://community.openai.com/u/WillEndure)\ **Replies:** 0\ **Last updated:** [November 27, 2025, 6:46pm UTC](https://community.openai.com/t/connecting-to-gpts-through-api-again/1367761 "2025-11-27T18:46:44Z")
This question was asked before, but the answers were not very satisfactory as the OP was also asking about web browsing, so lets try again… I created some GPTs with specialized knowledge, and overall I feel that they wo… --- ## [Finding tools/files too slow to be usable](https://community.openai.com/t/finding-tools-files-too-slow-to-be-usable/1367294)
**Author:** [@gerry.ha](https://community.openai.com/u/gerry.ha)\ **Replies:** 4\ **Last updated:** [November 26, 2025, 4:09pm UTC](https://community.openai.com/t/finding-tools-files-too-slow-to-be-usable/1367294 "2025-11-26T16:09:00Z")
I have been working on a project that makes use of the Responses API, specifically tools and files. Unfortunately I am finding it to be unacceptably SLOW. Note the “nano” models are not intelligent enough for my use case… --- ## [How to implement a real-time flow (voice/text) + vector store (file\_search) for fluid audio+text responses?](https://community.openai.com/t/how-to-implement-a-real-time-flow-voice-text-vector-store-file-search-for-fluid-audio-text-responses/1367522)
**Author:** [@Andrea\_Arredondo](https://community.openai.com/u/Andrea_Arredondo)\ **Replies:** 0\ **Last updated:** [November 25, 2025, 6:16pm UTC](https://community.openai.com/t/how-to-implement-a-real-time-flow-voice-text-vector-store-file-search-for-fluid-audio-text-responses/1367522 "2025-11-25T18:16:55Z")
Hello community, I am developing a chatbot that requires the following functionality: accept voice or text input, use a file search tool (vector store/uploaded documents) for context retrieval, and generate a fluid resp… --- ## [Vector Store File Attributes in Responses API file\_search not being used?](https://community.openai.com/t/vector-store-file-attributes-in-responses-api-file-search-not-being-used/1367014)
**Author:** [@Christopher\_Beltran](https://community.openai.com/u/Christopher_Beltran)\ **Replies:** 2\ **Last updated:** [November 21, 2025, 8:17am UTC](https://community.openai.com/t/vector-store-file-attributes-in-responses-api-file-search-not-being-used/1367014 "2025-11-21T08:17:42Z")
I’m using the Responses API and the file\_search tool to search within a given vector store. For some reason when I don’t specify filters against the attributes of the files things work great. If I specify filters again… --- ## [Deleted file remains in Assistant API response, causing “No such File object” error on update](https://community.openai.com/t/deleted-file-remains-in-assistant-api-response-causing-no-such-file-object-error-on-update/1363773)
**Author:** [@milad85](https://community.openai.com/u/milad85)\ **Replies:** 2\ **Last updated:** [November 19, 2025, 12:20am UTC](https://community.openai.com/t/deleted-file-remains-in-assistant-api-response-causing-no-such-file-object-error-on-update/1363773 "2025-11-19T00:20:03Z")
Hi everyone :waving\_hand:, I’ve encountered a blocking issue when managing files within a Vector Store linked to an Assistant. I’m creating a VectorStore and uploading files to it via the API. When I delete a file from… --- ## [Max\_num\_results limit decreased?](https://community.openai.com/t/max-num-results-limit-decreased/1365691)
**Author:** [@mcoster](https://community.openai.com/u/mcoster)\ **Replies:** 0\ **Last updated:** [November 9, 2025, 12:36pm UTC](https://community.openai.com/t/max-num-results-limit-decreased/1365691 "2025-11-09T12:36:24Z")
Hi, since a few days I’m getting en error on my vector store file search about the limit. To increase number os results, I had it set to 200 using the Responses API. But now I’m getting a tools.max\_num\_results error tell… --- ## [Vector Stores are not retrievable via list all](https://community.openai.com/t/vector-stores-are-not-retrievable-via-list-all/1359423)
**Author:** [@m.ackermann](https://community.openai.com/u/m.ackermann)\ **Replies:** 23\ **Last updated:** [November 4, 2025, 9:00am UTC](https://community.openai.com/t/vector-stores-are-not-retrievable-via-list-all/1359423 "2025-11-04T09:00:21Z")
Currently the list vector stores endpoint in the OpenAI API is not returning any vector stores. Getting the vector store directly via id is working fine, but using the list all will yield no results. I used the followin… --- ## [How does File Search work in the Responses API, and what is its pricing?](https://community.openai.com/t/how-does-file-search-work-in-the-responses-api-and-what-is-its-pricing/1363656)
**Author:** [@joao.petrini](https://community.openai.com/u/joao.petrini)\ **Replies:** 0\ **Last updated:** [October 23, 2025, 8:12pm UTC](https://community.openai.com/t/how-does-file-search-work-in-the-responses-api-and-what-is-its-pricing/1363656 "2025-10-23T20:12:01Z")
Hey guys, I created an agent using the Responses API with GPT-4.1 Mini, and it’s going well, but I noticed something strange in consumption: • Around 7k tokens per message • And 22k tokens per search on File Search + … --- ## [.md files not being accepted by file search](https://community.openai.com/t/md-files-not-being-accepted-by-file-search/1361602)
**Author:** [@tyron](https://community.openai.com/u/tyron)\ **Replies:** 0\ **Last updated:** [October 8, 2025, 7:07pm UTC](https://community.openai.com/t/md-files-not-being-accepted-by-file-search/1361602 "2025-10-08T19:07:54Z")
Previously, these exact same .md files were being accepted by the File Search tool in a playground prompt I was working on. However, the next day when I tried uploading these same files into a new prompt they were reject… --- ## [Access ResponseFileSearchToolCall queries before the search is done](https://community.openai.com/t/access-responsefilesearchtoolcall-queries-before-the-search-is-done/1360875)
**Author:** [@bsamuele](https://community.openai.com/u/bsamuele)\ **Replies:** 0\ **Last updated:** [October 3, 2025, 6:25am UTC](https://community.openai.com/t/access-responsefilesearchtoolcall-queries-before-the-search-is-done/1360875 "2025-10-03T06:25:39Z")
When streaming a response using the Responses API, we are able to access the queries used in a file search item only when we receive its corresponding ResponseOutputItemDoneEvent—at this point, it’s sufficient to do even… --- ## [Bug or wrong documentation: Responses api vector\_store\_ids array limited to one](https://community.openai.com/t/bug-or-wrong-documentation-responses-api-vector-store-ids-array-limited-to-one/1165564)
**Author:** [@dpiret1](https://community.openai.com/u/dpiret1)\ **Replies:** 6\ **Last updated:** [September 25, 2025, 1:19pm UTC](https://community.openai.com/t/bug-or-wrong-documentation-responses-api-vector-store-ids-array-limited-to-one/1165564 "2025-09-25T13:19:40Z")
According to the documentation, the file\_search tool of the Responses API admits several vector store ids: vector\_store\_ids array Required The IDs of the vector stores to search. But adding more than one throws an erro… --- ## [Best practice for using multiple documents with File Search in a single Responses API call?](https://community.openai.com/t/best-practice-for-using-multiple-documents-with-file-search-in-a-single-responses-api-call/1359687)
**Author:** [@Taci](https://community.openai.com/u/Taci)\ **Replies:** 0\ **Last updated:** [September 24, 2025, 12:44pm UTC](https://community.openai.com/t/best-practice-for-using-multiple-documents-with-file-search-in-a-single-responses-api-call/1359687 "2025-09-24T12:44:42Z")
TL;DR: I want the model (GPT-5) to, in one Responses API call, to search through 3 files, each for a distinct separate purpose. What’s the best way to implement this? One vector store with all three? Three vector stores,… --- ## [Questions about File Upload Limits with the OpenAI File API](https://community.openai.com/t/questions-about-file-upload-limits-with-the-openai-file-api/1358472)
**Author:** [@brother0326](https://community.openai.com/u/brother0326)\ **Replies:** 3\ **Last updated:** [September 22, 2025, 5:21pm UTC](https://community.openai.com/t/questions-about-file-upload-limits-with-the-openai-file-api/1358472 "2025-09-22T17:21:30Z")
Hi, I have a question regarding OpenAI file uploads. According to the File uploads FAQ, it seems that the number of files you can upload depends on your ChatGPT plan. However, when I use the OpenAI File API, I don’t se… --- ## [File Search Still Broken - Day 4 of Production Outage](https://community.openai.com/t/file-search-still-broken-day-4-of-production-outage/1357294)
**Author:** [@hsgaurav](https://community.openai.com/u/hsgaurav)\ **Replies:** 8\ **Last updated:** [September 22, 2025, 5:12pm UTC](https://community.openai.com/t/file-search-still-broken-day-4-of-production-outage/1357294 "2025-09-22T17:12:43Z")
It’s been 3+ days since this started (Sept 7th) and my production systems are still completely broken. What’s happening: Assistant API file search returns “I couldn’t find any specific sections in uploaded files” for… [Next page](https://community.openai.com/tag/file-search/707.md?match_all_tags=true&page=1&tags%5B%5D=file-search)