Revamp Agentic Inference benchmark guide - #431
Merged
hvagadia merged 3 commits intoJul 29, 2026
Merged
Conversation
Contributor
|
Caution The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased. |
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
hvagadia
marked this pull request as ready for review
July 29, 2026 19:12
Contributor
|
Caution The consumer version of Gemini Code Assist on GitHub has been sunset. All code review activity has officially ceased. |
nvzhihanj
approved these changes
Jul 29, 2026
nv-alicheng
approved these changes
Jul 29, 2026
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #431 +/- ##
=======================================
Coverage ? 81.58%
=======================================
Files ? 143
Lines ? 19163
Branches ? 0
=======================================
Hits ? 15635
Misses ? 3528
Partials ? 0 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
leopck
approved these changes
Jul 29, 2026
leopck
reviewed
Jul 29, 2026
leopck
reviewed
Jul 29, 2026
tianmu-li
reviewed
Jul 29, 2026
tianmu-li
approved these changes
Jul 29, 2026
arekay-nv
reviewed
Jul 29, 2026
Comment on lines
-171
to
+99
| Agentic inference benchmarks can have a long tail because different users receive | ||
| trajectories with very different turn counts, delays, and generated lengths. In | ||
| large runs this tail can last up to an hour after steady-state work has already | ||
| ended, so the benchmark separates the performance window from the remaining | ||
| accuracy/logging drain. | ||
| Agentic inference benchmarks can have a long tail because different users receive trajectories with very different turn counts, delays, and generated lengths. In large runs this tail can last up to an hour after steady-state work has already ended, so the benchmark separates the performance window from the remaining accuracy/logging drain. | ||
|
|
||
| The benchmark stops performance tracking when the first active user finishes its | ||
| final assigned trajectory. It emits `STOP_PERFORMANCE_TRACKING` at that point to | ||
| avoid measuring the tail. Turns issued before this event remain in the | ||
| performance window even if they finish later; turns issued after it are excluded | ||
| from performance metrics. | ||
| The benchmark stops performance tracking when the first active user finishes its final assigned trajectory. It emits `STOP_PERFORMANCE_TRACKING` at that point to avoid measuring the tail. Turns issued before this event remain in the performance window even if they finish later; turns issued after it are excluded from performance metrics. | ||
|
|
||
| For final submissions, keep | ||
| `agentic_inference.stop_issuing_on_first_user_complete: false` so the client finishes | ||
| already-started trajectories for accuracy. During optimization, set it to `true` | ||
| to stop issuing future turns at the performance boundary and shorten the tail. | ||
| For official submissions, submitters must set `agentic_inference.stop_issuing_on_first_user_complete` to `false` so the client finishes already-started trajectories for accuracy. During optimization, set it to `true` to stop issuing future turns at the performance boundary and shorten the tail. |
Collaborator
There was a problem hiding this comment.
These don't seem like real diffs - only formatting?
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
7 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.
Summary
Why
Submitters need a single, model-aware guide that distinguishes configurable deployment choices from the settings required for official Agentic Inference submissions.
Validation
uv run pre-commit run --all-filesBenchmarkConfig.from_yaml_file