[ML] Append all data to Chat Completion buffer#127658
Merged
prwhelan merged 8 commits intoelastic:mainfrom May 5, 2025
Merged
Conversation
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
prwhelan
added
>bug
:ml
Collaborator
|
Hi @prwhelan, I've created a changelog YAML for you. |
Collaborator
|
Pinging @elastic/ml-core (Team:ML) |
Contributor
jonathan-buttner
left a comment
jonathan-buttner
left a comment
There was a problem hiding this comment.
Looks good, left some questions.
| subscription.request(n); | ||
| } | ||
| } else { | ||
| downstream.onNext(new Results(DequeUtils.of(buffer.poll()))); |
Contributor
There was a problem hiding this comment.
Is there only 1 thread accessing the buffer? Or is there a chance that we could check for isEmpty() and then some other thread picks up the item before this thread?
Member
Author
There was a problem hiding this comment.
There is only 1 thread calling request, but we can be safe and change to bufer.poll() and check if it's null
| @Override | ||
| public void onComplete() {} | ||
| }); | ||
| assertThat(counter.get(), equalTo(2)); |
Contributor
There was a problem hiding this comment.
Just to make sure I understand, does this test that we only get a 1 result even if we have multiple in a single item?
Member
Author
There was a problem hiding this comment.
Yes because we'll only call onNext once per chunk, so if we send a chunk of 2 elements then counter will equal 1. Let me change to mockito spies so that's easier to read (I think)
jonathan-buttner
approved these changes
May 5, 2025
Collaborator
💔 Backport failed
You can use sqren/backport to manually backport by running |
ywangd
pushed a commit
to ywangd/elasticsearch
that referenced
this pull request
May 9, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
jfreden
pushed a commit
to jfreden/elasticsearch
that referenced
this pull request
May 12, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
prwhelan
added a commit
to prwhelan/elasticsearch
that referenced
this pull request
May 19, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
prwhelan
added a commit
to prwhelan/elasticsearch
that referenced
this pull request
May 19, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
elasticsearchmachine
pushed a commit
that referenced
this pull request
May 19, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
prwhelan
added a commit
that referenced
this pull request
May 19, 2025
Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
prwhelan
added a commit
to prwhelan/elasticsearch
that referenced
this pull request
May 19, 2025
…tic#128134) Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
elasticsearchmachine
pushed a commit
that referenced
this pull request
May 20, 2025
…128164) Moved the Chat Completion buffer into the StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer. This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer. This fixes a bug where onComplete would omit trailing items in the buffer.
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
4 participants
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.
Moved the Chat Completion buffer into the
StreamingUnifiedChatCompletionResults so that all Chat Completion responses can benefit from it. Chat Completions is meant to adhere to OpenAI as much as possible, and OpenAI only sends one response chunk at a time. All implementations of Chat Completions will now buffer.
This fixes a bug where more than two chunks in a single item would be dropped, instead they are all added to the buffer.
This fixes a bug where onComplete would omit trailing items in the buffer.