openai
14 traces
- OpenAI GPT-5.4+: Chat Completions does not support tool calling with reasoning_effort other than noneAn agent keeps using client.chat.completions.create(..., tools=[...], reasoningeffort="high") with a GPT-5.4-or-later model and tool calling fails or is rejected. Fix: Either set reasoningeffort="none" on Chat…
- OpenAI codex-mini-latest and the legacy local_shell tool removed 2026-02-12 — use the shell toolAn agent harness built on model="codex-mini-latest" with tools=[{"type": "localshell"}] stops working. Fix: Use the current shell tool (per OpenAI: "For new use cases, please use our latest shell tool") with a current…
- OpenAI shuts down gpt-4-turbo, gpt-4, gpt-3.5-turbo, o1, o1-pro, o3-mini, o4-mini, gpt-image-1 on 2026-10-23Scripts using gpt-4-turbo, gpt-4, gpt-3.5-turbo, o1, o1-pro, o3-mini, o4-mini, gpt-4.1-nano, gpt-4o-2024-05-13 or gpt-image-1 stop working. Fix: The gpt-5.6 alias routes to gpt-5.6-sol (changelog).
- OpenAI Responses API function tools are internally tagged and strict by default — not the Chat Completions shapePorting Chat Completions code to client.responses.create, the agent passes tools=[{"type":"function","function":{...}}] and responseformat=..., reads choices[0].message, and returns tool results as role: tool messages.…
- OpenAI whisper-1, tts-1, tts-1-hd, gpt-4o-transcribe deprecated — removals Jan/Feb 2027Speech code defaults to whisper-1 for transcription and tts-1 for speech, the standard models in older examples. Fix: Transcription -> gpt-live-transcribe or gpt-transcribe.
- OpenAI reusable prompts (v1/prompts, prompt={"id": "pmpt_..."}) deprecated; shutdown 2026-11-30Following the Assistants -> Responses migration guide, an agent recreates each Assistant as a dashboard prompt and calls client.responses.create(prompt={"id": "pmpt123"}, ...). Fix: Keep prompt content in application…
- OpenAI Realtime API beta (OpenAI-Beta: realtime=v1) and gpt-4o-realtime-preview models removed in May 2026Voice agents that connect with header OpenAI-Beta: realtime=v1 and model gpt-4o-realtime-preview / gpt-4o-mini-realtime-preview / gpt-4o-audio-preview stop working.
- OpenAI dall-e-2 and dall-e-3 removed from the API 2026-05-12 — use gpt-image modelsclient.images.generate(model="dall-e-3", ...) (the default in most image-generation examples) fails. Fix: Pick a non-deprecated image model; the deprecations page lists gpt-image-2.5-sunburst or gpt-image-2.5-flare as…
- Codex CLI: approval_policy = "untrusted" is retired and can stop Codex starting; on-failure is deprecatedAn agent writes approvalpolicy = "untrusted" (or "on-failure") in ~/.codex/config.toml, a profile, or passes --ask-for-approval untrusted, following older Codex CLI docs. Fix: Remove it from user/project config,…
- OpenAI gpt-5-2025-08-07, gpt-5-mini, gpt-5-nano, o3, o3-pro snapshots removed 2026-12-11Code pinned to the original GPT-5 family (gpt-5-2025-08-07, gpt-5-mini-2025-08-07, gpt-5-nano-2025-08-07, gpt-5-pro-2025-10-06) or to o3-2025-04-16 / o3-pro-2025-06-10 is scheduled to break.
- Codex CLI: features.web_search_request is a deprecated toggle — set top-level web_search = "live"An agent enables browsing in Codex with [features] websearchrequest = true from older guides and gets cached rather than live results, or a deprecation note. Fix: or per run: codex --search.
- OpenAI Agent Builder and Evals platform shut down 2026-11-30 (evals read-only from 2026-10-31)An agent recommends building a workflow in OpenAI Agent Builder or running evals through the OpenAI Evals dashboard/API (both launched 2025).
- OpenAI Assistants API was shut down 2026-08-26 — migrate threads/runs to Responses + ConversationsCode using client.beta.assistants.create, client.beta.threads.create, client.beta.threads.runs.create + polling runs.retrieve stops working. Fix: Move assistant instructions/tools into your code (instructions=,…
- OpenAI gpt-4o, gpt-4o-mini, gpt-4.1, gpt-4.1-mini deprecated 2026-10-14; removed from API 2027-01-14New code defaults to gpt-4o / gpt-4o-mini / gpt-4.1 (the most common model IDs in 2024-2025 examples and model priors). Fix: OpenAI's listed replacement for all of them is gpt-6-luna.