18 Releases published by 1 person

• v1.97.2

  published Jul 30, 2025

• v1.98.0

  published Jul 30, 2025

• v1.99.0

  published Aug 5, 2025

• v1.99.1

  published Aug 5, 2025

• v1.99.2

  published Aug 7, 2025

• v1.99.3

  published Aug 8, 2025

• v1.99.4

  published Aug 8, 2025

• v1.99.5

  published Aug 8, 2025

• v1.99.6

  published Aug 9, 2025

• v1.99.7

  published Aug 11, 2025

• v1.99.8

  published Aug 11, 2025

• v1.99.9

  published Aug 12, 2025

• v1.100.0

  published Aug 18, 2025

• v1.100.1

  published Aug 18, 2025

• v1.100.2

  published Aug 19, 2025

• v1.100.3

  published Aug 20, 2025

• v1.101.0

  published Aug 21, 2025

• v1.102.0

  published Aug 26, 2025

19 Pull requests merged by 2 people

• chore: bump inline-snapshot version to 0.28.0

  #2590 merged Aug 27, 2025

• release: 1.102.0

  #2586 merged Aug 26, 2025

• release: 1.101.0

  #2577 merged Aug 21, 2025

• release: 1.100.3

  #2570 merged Aug 20, 2025

• release: 1.100.2

  #2569 merged Aug 19, 2025

• release: 1.100.1

  #2565 merged Aug 18, 2025

• release: 1.100.0

  #2562 merged Aug 18, 2025

• release: 1.99.9

  #2555 merged Aug 12, 2025

• release: 1.99.8

  #2552 merged Aug 11, 2025

• release: 1.99.7

  #2551 merged Aug 11, 2025

• release: 1.99.6

  #2536 merged Aug 9, 2025

• release: 1.99.5

  #2534 merged Aug 8, 2025

• release: 1.99.4

  #2531 merged Aug 8, 2025

• release: 1.99.3

  #2522 merged Aug 7, 2025

• release: 1.99.2

  #2517 merged Aug 7, 2025

• release: 1.99.1

  #2509 merged Aug 5, 2025

• release: 1.99.0

  #2504 merged Aug 5, 2025

• release: 1.98.0

  #2503 merged Jul 30, 2025

• release: 1.97.2

  #2494 merged Jul 30, 2025

13 Pull requests opened by 9 people

• Fix optional logprobs field in ResponseTextDeltaEvent model

  #2500 opened Jul 29, 2025

• Doc: Update streaming examples to use gpt-4o instead of gpt-3.5-turbo-instruct

  #2506 opened Aug 4, 2025

• fix: Add automatic MIME type detection for file uploads

  #2520 opened Aug 7, 2025

• fix: preserve MCP session across tool calls

  #2526 opened Aug 7, 2025

• fix: correct jitter comment to reflect 25% scaling

  #2566 opened Aug 19, 2025

• feat(examples): add dynamic proxy forwarding example

  #2567 opened Aug 19, 2025

• Auth provider that leaves room for future challenge auth to be implemented

  #2571 opened Aug 20, 2025

• Simple auth provider limited to providing arbitrary sets of key/values for authentication

  #2573 opened Aug 20, 2025

• fix: remove Required on Optional field

  #2576 opened Aug 21, 2025

• feat: Add domain filtering options to web_search tool

  #2578 opened Aug 22, 2025

• fix: support azure new generation api version

  #2585 opened Aug 25, 2025

• Add support for token refresh by accepting a callable api_key parameter value

  #2588 opened Aug 26, 2025

• release: 1.102.1

  #2589 opened Aug 26, 2025

33 Issues closed by 9 people

• MCP Server Session Terminates After 1st Interaction

  #2513 closed Aug 27, 2025

• Lack of determinism when making API calls through the AzureOpenAI object versus vanilla curl requests with postman

  #2559 closed Aug 27, 2025

• client.chat.completions.create works with neither messages or input

  #2560 closed Aug 27, 2025

• `AsyncResponses.stream` is missing `conversation` parameter

  #2582 closed Aug 27, 2025

• background parameter is not being passed to create call in AsyncResponses client stream method

  #2579 closed Aug 27, 2025

• Add domain filtering options to web_search tool in Responses API

  #2572 closed Aug 27, 2025

• client.evals.create throws NameError

  #2574 closed Aug 27, 2025

• Add control message for Granite3.3

  #2583 closed Aug 25, 2025

• ImportError: cannot import name 'ResponseTextConfigParam' from 'openai.types.responses'

  #2564 closed Aug 19, 2025

• Built-in Streaming Response Buffer with Auto-Assembly

  #2558 closed Aug 13, 2025

• Pydantic model to appropiate schema for structured data

  #2557 closed Aug 12, 2025

• Random breaking change inserted by code bot

  #2553 closed Aug 11, 2025

• ChatCompletionMessageToolCallParam "cannot instantiate typing.Union" error in 1.99.4

  #2541 closed Aug 11, 2025

• Edge Case 17

  #2498 closed Aug 11, 2025

• ResponseReasoningSummaryDoneEvent missing from openai.types.responses in v1.99.0

  #2511 closed Aug 11, 2025

• Responses API with reused prompts and GPT-5

  #2540 closed Aug 11, 2025

• gpt-5-mini Returns Empty `output_text` with Responses API

  #2546 closed Aug 11, 2025

• Post `v1.99.2` breaking changes

  #2535 closed Aug 8, 2025

• responses.parse() fails with GPT-5 reasoning/verbosity parameters

  #2532 closed Aug 8, 2025

• Regression: response.output_text is broken since 1.99.2

  #2523 closed Aug 8, 2025

• Regression: ChatCompletionToolParam no longer a valid type in 1.99.2

  #2525 closed Aug 8, 2025

• Tool call seems problematic with openai sdk 1.99.3

  #2529 closed Aug 8, 2025

• Cannot use verbosity for GPT-5 API

  #2528 closed Aug 8, 2025

• —

  #2518 closed Aug 8, 2025

• output_text property was removed

  #2524 closed Aug 7, 2025

• `output_text` removed?

  #2521 closed Aug 7, 2025

• GPT 5 Let s GOOOO

  #2519 closed Aug 7, 2025

• Any plan to support gpt-oss's built-in tools?

  #2515 closed Aug 7, 2025

• Test failures with inline-snapshot>=0.25.0

  #2514 closed Aug 7, 2025

• How to customize the header？

  #2516 closed Aug 7, 2025

• Responses API broke in 1.99

  #2510 closed Aug 5, 2025

• Quickstart example.py ImportError

  #2508 closed Aug 5, 2025

• python sdk prompt_cache_key support?

  #2499 closed Jul 30, 2025

15 Issues opened by 13 people

• AzureOpenAi is unable to process chat.completions when the API version is set to preview.

  #2584 opened Aug 25, 2025

• `ImageGenerationCall` does not have all fields that API returns

  #2580 opened Aug 22, 2025

• Error "Item ‘rs_ABCD’ of type ‘reasoning’ was provided without its required..." when using CodeInterpreter

  #2561 opened Aug 15, 2025

• text_tokens and image_tokens not documented or typed

  #2554 opened Aug 11, 2025

• Add shell auto-completion support for OpenAI CLI (bash, zsh, fish, powershell)

  #2545 opened Aug 10, 2025

• usage.prompt_tokens_details=None

  #2544 opened Aug 10, 2025

• httpx.PoolTimeout occurs frequently when using SyncClient for multiple sequential requests

  #2539 opened Aug 8, 2025

• Unexpected Token Count Mismatch in tiktoken Integration with openai-python

  #2538 opened Aug 8, 2025

• GPT5 Models | Tool choices other than 'auto' are not supported with model

  #2537 opened Aug 8, 2025

• Schema validation error

  #2533 opened Aug 8, 2025

• DeltaToolCall only supports `function`, missing `custom` tool type

  #2530 opened Aug 8, 2025

• Update index documentation

  #2512 opened Aug 6, 2025

• Please add audio in streaming response for gpt-4o-audio-preview

  #2507 opened Aug 4, 2025

• Realtime API Pydantic Models incomplete for RealtimeResponseStatus type

  #2502 opened Jul 30, 2025

• Make ResponseOutputMessageParam.id genuinely optional and support message payloads without an OpenAI‑generated ID

  #2501 opened Jul 30, 2025

9 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Uploading PDF via Files API and using in Streaming gives 400 bad request

  #2472 commented on Aug 7, 2025 • 0 new comments

• Pydantic conversion logic for structured outputs is broken for models containing dictionaries

  #2004 commented on Aug 18, 2025 • 0 new comments

• Validation error for ResponseTextDeltaEvent after updating to 1.97.1

  #2489 commented on Aug 21, 2025 • 0 new comments

• At debug log level API requests to OpenAI get logged with all headers, including an API key, in plaintext

  #1196 commented on Aug 27, 2025 • 0 new comments

• Support for File Inputs In Azure OpenAI

  #2300 commented on Aug 28, 2025 • 0 new comments

• docs(client): fix incorrect comment about timeout calculation.

  #2252 commented on Aug 2, 2025 • 0 new comments

• Bearer token provider addition to openaiclient

  #2470 commented on Jul 29, 2025 • 0 new comments

• fix(beta): expose vector_stores in Beta client (fixes #2451)

  #2478 commented on Jul 31, 2025 • 0 new comments

• Remove extra newline in README

  #2492 commented on Aug 7, 2025 • 0 new comments