Repository navigation
ref(server-utils): Use shared stream state and endStreamSpan for Bedrock streams - #26192
Merged
Merged
Conversation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…ock streams Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit a236111. Configure here.
Contributor
size-limit report 📦Compared: f206027 (PR test merge using develop at 2e4231e) Against: 2e4231e (develop) · base run #65527 ✅ Baseline matches: this diff contains only this PR's changes.
|
Contributor
Optional E2E testsSome optional checks were not successful. They do not block the merge. 6 not successful, 51 successful. See the run.
Successful checks
|
…stream-state # Conflicts: # dev-packages/node-integration-tests/suites/aws-serverless/bedrock/test.ts
andreiborza
marked this pull request as ready for review
November 9, 2026 15:26
andreiborza
requested review from
JPeer264,
chargome,
isaacs,
mydea and
nicohrubec
and removed request for
a team
November 9, 2026 15:27
JPeer264
approved these changes
Nov 10, 2026
| @@ -291,6 +291,31 @@ describe('BedrockRuntimeServiceExtension token usage with prompt caching', () => | |||
| expect(usageAttributes(span, EXPECTED_USAGE)).toEqual(EXPECTED_USAGE); | |||
| }); | |||
|
|
|||
Member
There was a problem hiding this comment.
l: titan, llama and nova are not tested, we could add a test for each of those to check if they work. Feel free to change the city to Vienna.
Suggested change
| it.each([ | |
| { | |
| family: 'Llama', | |
| modelId: 'meta.llama3-8b-instruct-v1:0', | |
| chunks: [ | |
| { generation: 'Paris', prompt_token_count: 12, generation_token_count: 1, stop_reason: null }, | |
| { generation: '.', prompt_token_count: null, generation_token_count: 2, stop_reason: 'stop' }, | |
| ], | |
| expected: { | |
| [GEN_AI_USAGE_INPUT_TOKENS]: 12, | |
| [GEN_AI_USAGE_OUTPUT_TOKENS]: 2, | |
| [GEN_AI_USAGE_TOTAL_TOKENS]: 14, | |
| [GEN_AI_RESPONSE_FINISH_REASONS]: '["stop"]', | |
| }, | |
| }, | |
| { | |
| family: 'Titan', | |
| modelId: 'amazon.titan-text-express-v1', | |
| chunks: [ | |
| { outputText: 'Paris', totalOutputTextTokenCount: null, completionReason: null, inputTextTokenCount: 5 }, | |
| { outputText: '.', totalOutputTextTokenCount: 2, completionReason: 'FINISH', inputTextTokenCount: null }, | |
| ], | |
| expected: { | |
| [GEN_AI_USAGE_INPUT_TOKENS]: 5, | |
| [GEN_AI_USAGE_OUTPUT_TOKENS]: 2, | |
| [GEN_AI_USAGE_TOTAL_TOKENS]: 7, | |
| [GEN_AI_RESPONSE_FINISH_REASONS]: '["FINISH"]', | |
| }, | |
| }, | |
| { | |
| family: 'Nova', | |
| modelId: 'amazon.nova-lite-v1:0', | |
| chunks: [ | |
| { contentBlockDelta: { delta: { text: 'Paris' }, contentBlockIndex: 0 } }, | |
| { messageStop: { stopReason: 'end_turn' } }, | |
| { metadata: { usage: { inputTokens: 7, outputTokens: 2 }, metrics: {}, trace: {} } }, | |
| ], | |
| expected: { | |
| [GEN_AI_USAGE_INPUT_TOKENS]: 7, | |
| [GEN_AI_USAGE_OUTPUT_TOKENS]: 2, | |
| [GEN_AI_USAGE_TOTAL_TOKENS]: 9, | |
| [GEN_AI_RESPONSE_FINISH_REASONS]: '["end_turn"]', | |
| }, | |
| }, | |
| ])('records token usage and finish reasons for a $family stream', async ({ modelId, chunks, expected }) => { | |
| const span = startInactiveSpan({ name: 'bedrock' }); | |
| const encoder = new TextEncoder(); | |
| const data = { | |
| body: (async function* () { | |
| for (const chunk of chunks) { | |
| yield { chunk: { bytes: encoder.encode(JSON.stringify(chunk)) } }; | |
| } | |
| })(), | |
| }; | |
| respond(span, 'InvokeModelWithResponseStream', { modelId }, data); | |
| await drain(data.body); | |
| expect(spanToJSON(span).attributes).toMatchObject(expected); | |
| }); | |
Member
Author
There was a problem hiding this comment.
Added, with Kawasaki as the city.
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.
What
Bedrock ConverseStream and InvokeModelWithResponseStream spans now collect stream data into the shared stream state and end through
endStreamSpan, like the other AI integrations. Stream spans now also getgen_ai.response.streamingand a total token count for all model families.Why
The shared state is the base for recording Bedrock outputs in a later PR, and it makes Bedrock stream spans consistent with the other AI integrations.
Extracted from #26118.