# \#function-calling **URL:** https://community.openai.com/tag/function-calling/199.md [Latest](https://community.openai.com/latest.md) · [Categories](https://community.openai.com/categories.md) · [Tags](https://community.openai.com/tags.md) --- ## [Agent Builder - Local Function gerar\_faixa times out after 2m06s - Missing output exec](https://community.openai.com/t/agent-builder-local-function-gerar-faixa-times-out-after-2m06s-missing-output-exec/1400820)
**Author:** [@valdeci\_maciel](https://community.openai.com/u/valdeci_maciel)\ **Replies:** 0\ **Last updated:** [September 25, 2026, 11:30pm UTC](https://community.openai.com/t/agent-builder-local-function-gerar-faixa-times-out-after-2m06s-missing-output-exec/1400820 "2026-09-25T23:30:39Z")
How to connect my website backend to Agent Builder Local Function gerar\_faixa? Always Incomplete Hi team, My agent vmaciel studio on env ccarenv\_b64\_Y2NhcmVudl82YWI2ZmOGYzMjg4MTkxOGFkYmQxYTBIMjK3MmUxMg with model gpt-6… --- ## [Gpt-realtime-2.1: production tool invocation regression — model answers instead of calling available tools](https://community.openai.com/t/gpt-realtime-2-1-production-tool-invocation-regression-model-answers-instead-of-calling-available-tools/1398438)
**Author:** [@WebPlanning\_SIM-Ltd](https://community.openai.com/u/WebPlanning_SIM-Ltd)\ **Replies:** 2\ **Last updated:** [September 18, 2026, 9:40am UTC](https://community.openai.com/t/gpt-realtime-2-1-production-tool-invocation-regression-model-answers-instead-of-calling-available-tools/1398438 "2026-09-18T09:40:41Z")
We have just completed a controlled qualification of gpt-realtime-2.1 in our production voice agent. The test is strictly controlled: same application, same architecture, same instructions, same tools, same voice config… --- ## [Support for end of turn in gpt-live voice model](https://community.openai.com/t/support-for-end-of-turn-in-gpt-live-voice-model/1397759)
**Author:** [@anatRvz](https://community.openai.com/u/anatRvz)\ **Replies:** 0\ **Last updated:** [September 15, 2026, 9:47am UTC](https://community.openai.com/t/support-for-end-of-turn-in-gpt-live-voice-model/1397759 "2026-09-15T09:47:21Z")
Hey there, I’m migrating my product from realtime to live model and one thing I’m really missing is the ability to know when a turn from the voice model is over. More specifically, I’m having a very hard time understan… --- ## [Gpt-live-1: session.start fails with "Type is not JSON serializable: decimal.Decimal" if any tool schema contains a float](https://community.openai.com/t/gpt-live-1-session-start-fails-with-type-is-not-json-serializable-decimal-decimal-if-any-tool-schema-contains-a-float/1397313)
**Author:** [@jayzuccarelli](https://community.openai.com/u/jayzuccarelli)\ **Replies:** 0\ **Last updated:** [September 14, 2026, 12:52am UTC](https://community.openai.com/t/gpt-live-1-session-start-fails-with-type-is-not-json-serializable-decimal-decimal-if-any-tool-schema-contains-a-float/1397313 "2026-09-14T00:52:53Z")
Any float literal anywhere in a delegated tool schema makes session.start fail on gpt-live-1. The request JSON is valid; it appears to break server-side. Error returned: Invalid AVAS session\_data: Type is not JSON seri… --- ## [An error occurs when one dialog calls another dialog](https://community.openai.com/t/an-error-occurs-when-one-dialog-calls-another-dialog/1394230)
**Author:** [@guwango](https://community.openai.com/u/guwango)\ **Replies:** 2\ **Last updated:** [September 9, 2026, 11:21am UTC](https://community.openai.com/t/an-error-occurs-when-one-dialog-calls-another-dialog/1394230 "2026-09-09T11:21:57Z")
My scenario is that I have created different conversations under different project directories. I need a single conversation to call the conversations in these different directories in order to fulfill my requirements. H… --- ## [Feature Request: Proactive Personal Voice Companion — Allow ChatGPT to Initiate Voice Conversations](https://community.openai.com/t/feature-request-proactive-personal-voice-companion-allow-chatgpt-to-initiate-voice-conversations/1389836)
**Author:** [@deniss.cie](https://community.openai.com/u/deniss.cie)\ **Replies:** 1\ **Last updated:** [September 8, 2026, 12:29am UTC](https://community.openai.com/t/feature-request-proactive-personal-voice-companion-allow-chatgpt-to-initiate-voice-conversations/1389836 "2026-09-08T00:29:38Z")
Feature Request: Proactive Personal Voice Companion Idea Allow ChatGPT to proactively contact a user — including initiating a Voice conversation — based on user-defined permissions, schedules, context, and recent activit… --- ## [Generate images directly while we’re on a voice call,](https://community.openai.com/t/generate-images-directly-while-we-re-on-a-voice-call/1395382)
**Author:** [@michitucu](https://community.openai.com/u/michitucu)\ **Replies:** 0\ **Last updated:** [September 7, 2026, 1:43am UTC](https://community.openai.com/t/generate-images-directly-while-we-re-on-a-voice-call/1395382 "2026-09-07T01:43:18Z")
Estaría buenísimo poder generar imágenes directamente mientras estamos en una llamada de voz, sin tener que salir del modo voz y pasar al chat de texto. Por ejemplo, si durante una conversación le pido a ChatGPT que gen… --- ## [Recommended handling for outcome-unknown external actions after dispatch?](https://community.openai.com/t/recommended-handling-for-outcome-unknown-external-actions-after-dispatch/1394602)
**Author:** [@sdh\_own\_standard](https://community.openai.com/u/sdh_own_standard)\ **Replies:** 1\ **Last updated:** [September 3, 2026, 2:22pm UTC](https://community.openai.com/t/recommended-handling-for-outcome-unknown-external-actions-after-dispatch/1394602 "2026-09-03T14:22:45Z")
Hello — I have one non-confidential architecture question about agents performing external actions. For a non-idempotent tool call or real-world action, how should an application handle the case where the action has alr… --- ## [\[Bug\] gpt-5.6-sol returns HTTP 500 after shell + custom function continuation](https://community.openai.com/t/bug-gpt-5-6-sol-returns-http-500-after-shell-custom-function-continuation/1387737)
**Author:** [@jhariani](https://community.openai.com/u/jhariani)\ **Replies:** 1\ **Last updated:** [July 23, 2026, 9:19pm UTC](https://community.openai.com/t/bug-gpt-5-6-sol-returns-http-500-after-shell-custom-function-continuation/1387737 "2026-07-23T21:19:20Z")
I am seeing a reproducible HTTP 500 with gpt-5.6-sol on the Responses API when a response uses the hosted shell tool and then calls a developer-defined function. The initial request completes successfully and returns th… --- ## [\[Bug\] gpt-5.6-luna intermittently returns HTTP 500 for image-only function outputs](https://community.openai.com/t/bug-gpt-5-6-luna-intermittently-returns-http-500-for-image-only-function-outputs/1386679)
**Author:** [@ido](https://community.openai.com/u/ido)\ **Replies:** 7\ **Last updated:** [July 19, 2026, 9:49am UTC](https://community.openai.com/t/bug-gpt-5-6-luna-intermittently-returns-http-500-for-image-only-function-outputs/1386679 "2026-07-19T09:49:01Z")
I am seeing a reproducible issue with gpt-5.6-luna on the Responses API. We use the computer tool alongside a custom function. After the model calls the function, we continue with the same previous\_response\_id and retur… --- ## [Error in function calling in node.js Examples](https://community.openai.com/t/error-in-function-calling-in-node-js-examples/1356129)
**Author:** [@guptaa.pavan](https://community.openai.com/u/guptaa.pavan)\ **Replies:** 3\ **Last updated:** [March 23, 2026, 1:55pm UTC](https://community.openai.com/t/error-in-function-calling-in-node-js-examples/1356129 "2026-03-23T13:55:14Z")
Hello, The example provided for (Complete tool calling example) under javascript has errors in this link https://platform.openai.com/docs/guides/function-calling This is the correct code \`\`\` // 1. Define a list of ca… --- ## [GPT-5.4 Release, weird tool behaviour](https://community.openai.com/t/gpt-5-4-release-weird-tool-behaviour/1375860)
**Author:** [@jorge\_paiva](https://community.openai.com/u/jorge_paiva)\ **Replies:** 8\ **Last updated:** [March 10, 2026, 2:23pm UTC](https://community.openai.com/t/gpt-5-4-release-weird-tool-behaviour/1375860 "2026-03-10T14:23:20Z")
I’m currently building an application using the Responses API. I’m using GPT-5.2 Since the GPT-5.4 release, I’ve been experiencing issues when enabling our internal function tools alongside the built-in tools ("type": … --- ## [Function calling with recursive variant in an \`anyOf\` erroneously collapses to the simplest schema-valid variant](https://community.openai.com/t/function-calling-with-recursive-variant-in-an-anyof-erroneously-collapses-to-the-simplest-schema-valid-variant/1372550)
**Author:** [@krt1224](https://community.openai.com/u/krt1224)\ **Replies:** 0\ **Last updated:** [January 26, 2026, 7:25am UTC](https://community.openai.com/t/function-calling-with-recursive-variant-in-an-anyof-erroneously-collapses-to-the-simplest-schema-valid-variant/1372550 "2026-01-26T07:25:14Z")
I’ve noticed a degenerate pattern in my function calling, which I will illustrate with a minimal example. I’ve found that when I supply the model with an anyOf, where the model might need to select one of multiple comple… --- ## [Has anyone managed to get a tool\_call working when stream=True?](https://community.openai.com/t/has-anyone-managed-to-get-a-tool-call-working-when-stream-true/498867)
**Author:** [@Xeniox](https://community.openai.com/u/Xeniox)\ **Replies:** 23\ **Last updated:** [January 23, 2026, 12:45am UTC](https://community.openai.com/t/has-anyone-managed-to-get-a-tool-call-working-when-stream-true/498867 "2026-01-23T00:45:48Z")
Hi there! I’m new here so please forgive any poor choices etc. I have been playing around with the OpenAI API for a few months now, and this is how i previously handled function calls and streaming in python: if chunk\["… --- ## [How to Manage Transient Tool Results](https://community.openai.com/t/how-to-manage-transient-tool-results/1371680)
**Author:** [@wnmills3](https://community.openai.com/u/wnmills3)\ **Replies:** 2\ **Last updated:** [January 14, 2026, 7:44pm UTC](https://community.openai.com/t/how-to-manage-transient-tool-results/1371680 "2026-01-14T19:44:50Z")
I’ve implemented a couple of tools, and I’m getting inconsistent results across successive calls to retrieve transient results. The simplest is get\_timeofday, which returns the current datetime information. By definition… --- ## [Documentation fix to code for Function calling](https://community.openai.com/t/documentation-fix-to-code-for-function-calling/1371391)
**Author:** [@matt\_terial](https://community.openai.com/u/matt_terial)\ **Replies:** 5\ **Last updated:** [January 10, 2026, 8:42pm UTC](https://community.openai.com/t/documentation-fix-to-code-for-function-calling/1371391 "2026-01-10T20:42:23Z")
Hi I noticed a code example in the docs that produces errors and does not run successfully. Is there a way I can provide fixes to help future devs? https://platform.openai.com/docs/guides/function-calling#function-tool-… --- ## [Is Sequential Tool Calling possible?](https://community.openai.com/t/is-sequential-tool-calling-possible/1369938)
**Author:** [@lola\_emil](https://community.openai.com/u/lola_emil)\ **Replies:** 2\ **Last updated:** [January 9, 2026, 11:05am UTC](https://community.openai.com/t/is-sequential-tool-calling-possible/1369938 "2026-01-09T11:05:41Z")
Hi, I was wondering if sequential tool calling is possible. I have these 2 functions below: $tools = \[ \[ 'name' =\> 'user\_attendance\_data', 'type' =\> 'function', 'function' =\> \[ 'n… --- ## [Realtime API Function Calling](https://community.openai.com/t/realtime-api-function-calling/1369687)
**Author:** [@chinmay1](https://community.openai.com/u/chinmay1)\ **Replies:** 1\ **Last updated:** [December 19, 2025, 6:50am UTC](https://community.openai.com/t/realtime-api-function-calling/1369687 "2025-12-19T06:50:34Z")
My prompt requires the Realtime API to call “irrelevant question” tool when a user asks an irrelevant question. We count the number of times a user has asked such questions and warn them the first two times. The third t… --- ## [Editing function\_call.arguments in Agents SDK Has No Effect — How to Reflect Updated Form State?](https://community.openai.com/t/editing-function-call-arguments-in-agents-sdk-has-no-effect-how-to-reflect-updated-form-state/1368904)
**Author:** [@Cemalettin\_Yilmaz](https://community.openai.com/u/Cemalettin_Yilmaz)\ **Replies:** 0\ **Last updated:** [December 10, 2025, 10:39am UTC](https://community.openai.com/t/editing-function-call-arguments-in-agents-sdk-has-no-effect-how-to-reflect-updated-form-state/1368904 "2025-12-10T10:39:34Z")
Hi everyone — I’m using the OpenAI Agents SDK (Python) and I’m trying to “rehydrate” a chat from my DB by feeding Runner.run() the previous run items from result.to\_input\_list(). I noticed something that feels like the … --- ## [Typescript Sample Code for Function Calling is wrong](https://community.openai.com/t/typescript-sample-code-for-function-calling-is-wrong/1368645)
**Author:** [@goyalkavya](https://community.openai.com/u/goyalkavya)\ **Replies:** 1\ **Last updated:** [December 7, 2025, 7:11pm UTC](https://community.openai.com/t/typescript-sample-code-for-function-calling-is-wrong/1368645 "2025-12-07T19:11:24Z")
The typescript sample code present here is wrong, The input defined and given as message to the LLM calls is named as input. However, the results of the tool-execution are added to an undefined variable named input\_list … --- ## [Inconsistent tool calling on GPT-4o & GPT-4.1](https://community.openai.com/t/inconsistent-tool-calling-on-gpt-4o-gpt-4-1/1366299)
**Author:** [@liger](https://community.openai.com/u/liger)\ **Replies:** 0\ **Last updated:** [November 14, 2025, 1:55pm UTC](https://community.openai.com/t/inconsistent-tool-calling-on-gpt-4o-gpt-4-1/1366299 "2025-11-14T13:55:17Z")
Hi, I have question about tool calling and consistency when using different models. I noticed some models like GPT-4o almost always call the provided (single) tool multiple times depending on the user prompt and other m… --- ## [Need Help: "No tool output found for function call" Error After Function Call in Responses API](https://community.openai.com/t/need-help-no-tool-output-found-for-function-call-error-after-function-call-in-responses-api/1245970)
**Author:** [@binjan.iyer](https://community.openai.com/u/binjan.iyer)\ **Replies:** 5\ **Last updated:** [October 28, 2025, 5:31pm UTC](https://community.openai.com/t/need-help-no-tool-output-found-for-function-call-error-after-function-call-in-responses-api/1245970 "2025-10-28T17:31:31Z")
Hi all, I’m working with the Responses API for my project and I’m encountering an issue after a function call. Here’s how I’ve set up my tool definitions: tools = \[ { "type": "web\_search\_preview" }, { … --- ## [Realtime tools: enum not enforced; invalid strings slip through](https://community.openai.com/t/realtime-tools-enum-not-enforced-invalid-strings-slip-through/1362990)
**Author:** [@zollipaul](https://community.openai.com/u/zollipaul)\ **Replies:** 2\ **Last updated:** [October 19, 2025, 3:10pm UTC](https://community.openai.com/t/realtime-tools-enum-not-enforced-invalid-strings-slip-through/1362990 "2025-10-19T15:10:48Z")
Issue Realtime api tool calling emits out-of-set strings for tool arguments even with strict: true and a JSON-Schema enum. Symptom Server validation fails with invalid\_enum\_value on the enum field. Expected Argument… --- ## [Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study](https://community.openai.com/t/development-of-a-conversational-artificial-intelligence-based-web-application-for-medical-consultations-prototype-study/1362717)
**Author:** [@jorgeguerrabrazil](https://community.openai.com/u/jorgeguerrabrazil)\ **Replies:** 0\ **Last updated:** [October 16, 2025, 9:59am UTC](https://community.openai.com/t/development-of-a-conversational-artificial-intelligence-based-web-application-for-medical-consultations-prototype-study/1362717 "2025-10-16T09:59:38Z")
Just sharing what I was able to do with the community: Development of a Conversational Artificial Intelligence–Based Web Application for Medical Consultations: Prototype Study It took more than one year to get published… --- ## [AI Chess Database — A sandbox for integrating AI into real applications](https://community.openai.com/t/ai-chess-database-a-sandbox-for-integrating-ai-into-real-applications/1362261)
**Author:** [@mdkusz](https://community.openai.com/u/mdkusz)\ **Replies:** 0\ **Last updated:** [October 13, 2025, 7:02am UTC](https://community.openai.com/t/ai-chess-database-a-sandbox-for-integrating-ai-into-real-applications/1362261 "2025-10-13T07:02:14Z")
Hi everyone, I’ve been following this community for a while and finally decided to share my first experiment integrating AI into a real (Windows) desktop application. I’ve recently started a long-term project to explor… --- ## [How can I prevent function call output from being saved in Conversation?](https://community.openai.com/t/how-can-i-prevent-function-call-output-from-being-saved-in-conversation/1358380)
**Author:** [@ronnie.ho](https://community.openai.com/u/ronnie.ho)\ **Replies:** 0\ **Last updated:** [September 16, 2025, 12:05pm UTC](https://community.openai.com/t/how-can-i-prevent-function-call-output-from-being-saved-in-conversation/1358380 "2025-09-16T12:05:59Z")
When I create a response to reply to a function call (type = function\_call\_output), how do I prevent this output from being saved into the conversation? I tried to set store = false, but if I do that, I don’t even get th… --- ## [Function calling schema validation bug: array parameters with prefixItems rejected (missing items error)](https://community.openai.com/t/function-calling-schema-validation-bug-array-parameters-with-prefixitems-rejected-missing-items-error/1357778)
**Author:** [@haskely](https://community.openai.com/u/haskely)\ **Replies:** 1\ **Last updated:** [September 12, 2025, 1:31pm UTC](https://community.openai.com/t/function-calling-schema-validation-bug-array-parameters-with-prefixitems-rejected-missing-items-error/1357778 "2025-09-12T13:31:13Z")
Problem When defining a function tool with an array parameter that uses prefixItems (valid JSON Schema 2020-12 syntax), the OpenAI API rejects the request with a 400 Bad Request error. Example request (shortened for cl… --- ## [New API changes on Chat Completion broke some previous working stuff](https://community.openai.com/t/new-api-changes-on-chat-completion-broke-some-previous-working-stuff/504311)
**Author:** [@aurimasniekis](https://community.openai.com/u/aurimasniekis)\ **Replies:** 6\ **Last updated:** [September 7, 2025, 10:06am UTC](https://community.openai.com/t/new-api-changes-on-chat-completion-broke-some-previous-working-stuff/504311 "2025-09-07T10:06:08Z")
Dear OpenAI Community, I’m writing to share and seek advice on a specific issue we’ve encountered with the recent changes to the Chat Completion API, which has significantly affected our application’s functionality. Ba… --- ## [\`FunctionTool\` runtime behavior contradicts OpenAPI spec](https://community.openai.com/t/functiontool-runtime-behavior-contradicts-openapi-spec/1355938)
**Author:** [@french](https://community.openai.com/u/french)\ **Replies:** 0\ **Last updated:** [September 3, 2025, 2:54am UTC](https://community.openai.com/t/functiontool-runtime-behavior-contradicts-openapi-spec/1355938 "2025-09-03T02:54:41Z")
Observed Behavior The most recent version of the OpenAI spec\[1\] claims that strict defaults to true when passed as strict: null to the Responses API, but the implementation actually appears to default it to false.\[2\] … --- ## [GPT-4.1 sometimes says “One moment…” / “Hang tight…” but never calls a tool](https://community.openai.com/t/gpt-4-1-sometimes-says-one-moment-hang-tight-but-never-calls-a-tool/1354480)
**Author:** [@dadonik](https://community.openai.com/u/dadonik)\ **Replies:** 6\ **Last updated:** [August 31, 2025, 3:30pm UTC](https://community.openai.com/t/gpt-4-1-sometimes-says-one-moment-hang-tight-but-never-calls-a-tool/1354480 "2025-08-31T15:30:51Z")
Hi all, I’m running into a strange behavior with the gpt-4.1-2025-04-14 model and wanted to check if others have seen this too. We use both the Chat Completions API and the Responses API, with ~10 different tools confi… [Next page](https://community.openai.com/tag/function-calling/199.md?match_all_tags=true&page=1&tags%5B%5D=function-calling)