test: add long streaming output chunk-size tests (21_01 / 21_02)#34
Open
huahua-ctrl wants to merge 1 commit into
Open
test: add long streaming output chunk-size tests (21_01 / 21_02)#34huahua-ctrl wants to merge 1 commit into
huahua-ctrl wants to merge 1 commit into
Conversation
Add module 21 (long_output) with two streaming cases that check SSE
chunk-size discipline for long outputs:
- 21_01: long content stream (delta.content chunks)
- 21_02: long tool_call arguments stream (delta.tool_calls[].function.arguments fragments)
Rules per case:
- Every chunk EXCLUDING the tail: len >= 10
- Every chunk INCLUDING the tail: len <= 300 (hard cap)
- Fraction of ALL chunks (INCLUDING the tail) with len < 200: > 90%
Any violation fails the case.
Fleet run against 10 direct M3 providers:
- 21_01 passes 4/10 (fireworks + together x3);
fails split between "per-token stream" (inferact x2 / magikcloud /
modular emit lots of <10c chunks) and "soft-cap ratio drop" (M3-test
at 84%)
- 21_02 passes 0/10; failures come in three flavors:
* value packed into one 800-902c fragment exceeding the hard cap
(M3-test / fireworks / fireworks-fp8 / modular)
* arguments not streamed incrementally at all — only 1 fragment
total (together x3 / inferact x2)
* mixed short + oversized fragments (magikcloud)
Co-Authored-By: Claude <noreply@anthropic.com>
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add module 21 (long_output) with two streaming cases that check SSE chunk-size discipline for long outputs:
Rules per case: