openai[minor]: add image generation to responses api #31424
Conversation
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎ 1 Skipped Deployment
|
CodSpeed Walltime Performance ReportMerging #31424 will not alter performanceComparing
|
CodSpeed Instrumentation Performance ReportMerging #31424 will not alter performanceComparing Summary
|
dosubot
bot
added
the
size:M
| @@ -118,6 +118,15 @@ | |||
|
|
|||
| _FUNCTION_CALL_IDS_MAP_KEY = "__openai_function_call_ids__" | |||
|
|
|||
| WellKnownTools = ( | |||
There was a problem hiding this comment.
should we import this from core
There was a problem hiding this comment.
Not at the moment. We should do something in the reverse direction likely -- core should not know anything specific about the responses API. We should resolve in a separate refactor to figure out how to decouple core completely from openai specifics.
|
|
||
| if tool["type"] == "image_generation" and "partial_images" in tool: | ||
| raise NotImplementedError( | ||
| "Partial image generation is not yet supported via the LangChain ChatOpenAI client. " |
There was a problem hiding this comment.
mild preference for dropping partial_images and emitting a warning.
alternatively, could update the error message here to make it very obvious what to do (drop partial_images from the image_generation tool definition)
There was a problem hiding this comment.
Updated the doc-string
dosubot
bot
added
size:L
Does not support partial images during generation at the moment. Before doing that I'd like to figure out how to specify the aggregation logic without requiring changes in core.