Repository navigation
feat: OpenAI Responses API for the openai model provider - #2196
Merged
Merged
Conversation
johanandren
approved these changes
Nov 9, 2026
johanandren
left a comment
Contributor
There was a problem hiding this comment.
LGTM once runtime is out
🤖 Review
Nits
- The
max-completion-tokensrow in the docs and thereference.confcomment don't say it is the fallback limit withapi = "responses"(the javadoc does). - The
reasoning-effortcomment inreference.confdoesn't say it applies to both APIs (the docs do).
octonato
approved these changes
Nov 9, 2026
patriknw
force-pushed
the
wip-openai-responses-api-patriknw
branch
from
November 9, 2026 15:46
8dc6d9e to
8f672d4
Compare
Contributor
Author
|
incorporated feedback, and rebased |
Contributor
Author
|
For the record, I tried this with a few samples and real openai gpt-5 responses api. Seems to be working! |
This was referenced Nov 12, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Requires akka-runtime https://github.com/lightbend/akka-runtime/pull/6090
🤖
The
openaimodel provider gains two settings:api("chat-completions" or "responses") andreasoning-effort. You can set both in configuration or onModelProvider.OpenAithrough the newOpenAiApienum and thewithApiandwithReasoningEffortmethods. The SDK passes both settings to the runtime, which calls the OpenAI Responses API whenapiis "responses". On that API the runtime asks OpenAI not to store the response, and sends the encrypted reasoning of a reasoning model back with the next call. The documentation describes whatthinking,reasoning-effortand the token limits do on each API, and it includes a compiled example configuration.The defaults are "chat-completions" and an empty reasoning effort. An application that sets neither key gets the same SPI provider as before, and a test checks this.