Repository navigation
Conversation
Signed-off-by: Daneyon Hansen <daneyon.hansen@solo.io>
Preserve historical profiles and validate one-round paired reports. Honor anonymous model downloads and correct repetition guidance. Signed-off-by: Daneyon Hansen <daneyon.hansen@solo.io>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds an experimental
praxis-standalonetreatment with a pinned custom-build recipe, chart overrides, runtime image checks, configurable GKE treatment/report pairs, and comparison metadata validation. A separate agentgateway v1.5.0 reference profile preserves the historical serving-engine/EPP settings. Campaigns now honor anonymous model downloads, and the README correctly describes repetition IDs.Praxis requires
full,llmd-ext-proc. Its integration cannot forward response bodies or trailers to the EPP; the guide requires real-EPP lifecycle qualification before interpreting GPU results. Reports distinguish simulator measurements from GPU inference and describe Service balancing as connection-based.Validation: scenario/runtime and reporting tests, actual Helm rendering against router v0.9.0, campaign/smoke orchestration tests, and ShellCheck passed. Added checks cover historical/current release guards and exactly three treatments producing the two selected comparisons. Native Praxis build, real-EPP qualification, and GPU measurements remain pending; no comparative results are included.