<!-- agent-source-envelope -->
# Open Kernel global-expert MoE block sizing | PTLLM

> Use the global expert count for routing density; do not confuse that heuristic with local expert IDs used by alignment. Retained outcome: worked with caveat.

Canonical: https://posttrainllm.com/experiments/openkernel-moe-20261023
<!-- /agent-source-envelope -->

Experiment · runtime perf

Use the global expert count for routing density; do not confuse that heuristic with local expert IDs used by alignment. Retained outcome: worked with caveat.

**Outcome:** worked with caveat

**Family:** runtime perf

**Evidence confidence:** exact

**Record:** infrastructure

## Evidence readout

2026-10-23 H200 scored run 84fb1b656d3c44e2: 1.0645x aggregate, reported 95% interval 1.0485-1.0804x; correctness passed with no flags.

## What the attempt taught

Use the global expert count for routing density; do not confuse that heuristic with local expert IDs used by alignment.

## Why it stopped or stayed bounded

The gain is specific to the supplied expert-parallel routing distribution and organizer H200 workload.

## Method and scope

- Experiment family: runtime perf.

- Record kind: infrastructure.

## Disposition

Experiment closed after scored run beb82b4224554523 finished at rank 283/749, outside the owner top-100 target. Retain the research artifact; no further jobs.

This retained disposition is historical evidence, not authorization to restart the experiment. A new run needs a fresh question, frozen evaluator, explicit resource budget, and scoped tracking issue.

## How to read this dossier

This canonical page separates the retained repository decision from the material that informed it. A source may describe an external claim, historical measurement, local observation, or planned procedure; those evidence classes are not interchangeable. Follow the provenance links before reusing a number or method.

Status and disposition describe what PostTrainLLM retained when this record was indexed. They are not a live product promise, a newly run benchmark, or permission to restart historical work. Unknown or unmeasured fields remain unknown rather than being treated as zero.

To reuse the record, first name the exact claim you need and trace it to the linked source. Then check whether the original environment, model revision, data split, evaluator, hardware, and budget match the proposed use. If they do not, treat the record as a hypothesis or design reference and run the smallest fresh comparison that can falsify it. Preserve negative outcomes and regressions beside any improvement; a local win on one slice does not silently become a general capability claim.

## Source provenance

The normalized record comes from `docs/attempts.json`. The links below are the tracked evidence and explanatory sources preserved with the record.

- [docs/learn/openkernel.md](https://posttrainllm.com/docs/learn/openkernel)

- [experiments/openkernel/results.json](https://github.com/PostTrainLLM/posttrainllm/blob/main/experiments/openkernel/results.json)

- [experiments/openkernel/muse_moe_w4a16/submission.py](https://github.com/PostTrainLLM/posttrainllm/blob/main/experiments/openkernel/muse_moe_w4a16/submission.py)

## Related evidence

- [Primary proof](https://posttrainllm.com/docs/learn/openkernel)

- [Applicable recipe](https://posttrainllm.com/recipes)

- [Learning path](https://posttrainllm.com/learn/paths/runtime-and-agents)

Canonical record: `experiment:openkernel-moe-20261023`