---
title: '56875'
pageLayout: default
slug: thoughts/inference/vllm-prs/56875
permalink: https://aarnphm.xyz/thoughts/inference/vllm-prs/56875.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
# [#56875](https://github.com/aarnphm/aarnphm.github.io/issues/56875) \[Perf\] Enable Kimi K3 shared expert sharding with DeepEP v2

[vllm-project/vllm#56875](https://github.com/vllm-project/vllm/pull/56875) · elvircrn · MERGED · created 2026-09-14T17:24:03Z · merged 2026-10-08T16:42:16Z

Signed-off-by: Elvir Crncevic <elvircrn@gmail.com>

## Purpose

Enable Kimi K3 shared experts sharing with DeepEP V2, previously only enabled for MegaMoE.

## Test Plan

## Test Result

---

<details>
<summary> Essential Elements of an Effective PR Description Checklist </summary>

- [ ] The purpose of the PR, such as “Fix some issue (link existing issues this PR will resolve)”.
- [ ] The test plan, such as providing test command.
- [ ] The test results, such as pasting the results comparison before and after, or e2e results
- [ ] (Optional) The necessary documentation update, such as updating `supported_models.md` and `examples` for a new model.

</details>

---

## comments

**mgoin** (2026-10-07T19:36:36Z):

/ci run

**github-actions** (2026-10-07T19:36:58Z):

✅ Triggered [Buildkite CI #93396](https://buildkite.com/vllm/ci/builds/93396) for commit `a14d7f0c2b59`.

**mgoin** (2026-10-07T19:37:02Z):

Can you share an eval/performance to validate the benefit?

**github-actions** (2026-10-07T19:38:24Z):

✅ @elvircrn, CI is now available for this PR.

- `/ci run` starts upstream CI; `/amd-ci run` starts AMD CI only.
- Your branch must contain every commit currently on its upstream target branch. Merge or rebase onto the latest target branch, then rerun the command. Append `--allow-stale` to a run command to test an outdated branch at your own risk.
- `/ci retry` retries failed jobs in the CI build for the current PR head. If the current head has no CI build, it starts a new CI build for the current head containing only jobs that failed in the latest earlier CI build for this PR.
- `/amd-ci retry` retries failed jobs in AMD CI for the current PR head. Use `/amd-ci run` when the current head has no AMD CI build.
- `/ci cancel` cancels scheduled or running CI builds for this PR branch; `/amd-ci cancel` does the same for AMD CI only.

<!-- vllm-ci-authorized -->

**vllm-agent** (2026-10-07T19:55:04Z):

<!-- ci-selector-shadow -->

### CI selector (shadow): 8 test steps (32 jobs) instead of 36 (52 jobs)

Shadow mode: this changes nothing about what CI runs. It shows what the evidence-based selector would pick for this PR, next to today’s rules. [How it works](https://github.com/vllm-project/ci-infra/tree/main/buildkite/ci_selector).

**Feedback welcome:** reply here if it would skip a step this change needs, or runs something unrelated.

| steps (jobs)           | Today’s rules | Selector | Would skip | Would add |
| ---------------------- | ------------- | -------- | ---------- | --------- |
| NVIDIA, CPU and others | 36 (52)       | 8 (32)   | 32 (42)    | 4 (22)    |
| AMD mirrors            | 33 (43)       | 1 (1)    | 32 (42)    | 0 (0)     |

<details><summary>Selector would run (8)</summary>

- `arm-cpu-test` ×3
- `basic-models-tests-extra-initialization` ×14
- `cpu-language-generation-and-pooling-model-tests` ×3
- `cpu-multi-modal-model-tests-n` ×4
- `cpu-spec-decode-tests`
- `kimi-k3-prefix-cache-4xb200` ×2
- `kimi-k3-unit-tests-b200`
- `multi-modal-processor` ×4

</details>

<details><summary>Would skip (today's rules run them) (32)</summary>

- `ascend-npu-test`
- `basic-correctness` ×2
- `basic-correctness-cpu-offload`
- `basic-correctness-cumem`
- `basic-correctness-prefetch-offload`
- `basic-correctness-sleep-mode`
- `basic-models-test-other-cpu`
- `basic-models-tests-initialization`
- `basic-models-tests-other`
- `benchmarks-cli-test`
- `distributed-model-tests-2-gpus` ×3
- `entrypoints-integration-api-server` ×4
- `entrypoints-integration-api-server-generate`
- `entrypoints-integration-api-server-openai-chat_completion`
- `entrypoints-integration-api-server-openai-completion`
- `entrypoints-integration-llm`
- `entrypoints-integration-multimodal`
- `entrypoints-integration-pooling`
- `entrypoints-integration-responses-api`
- `entrypoints-integration-speech_to_text`
- `kernels-fla-ops-test-b200`
- `kernels-mhc-test-b200`
- `kernels-root-misc-test-b200`
- `language-models-tests-granite-l4-compatibility`
- `language-models-tests-hybrid` ×2
- `language-models-tests-standard`
- `multi-modal-models-standard-1-qwen2`
- `multi-modal-models-standard-2-qwen3-gemma`
- `multi-modal-models-standard-3-llava-qwen2-vl`
- `multi-modal-models-standard-4-other-whisper`
- `multi-modal-processor-cpu` ×4
- `rl-entrypoints-tests`

</details>

<details><summary>Would add (today's rules do not run them) (4)</summary>

- `arm-cpu-test` ×3 (code map)
- `basic-models-tests-extra-initialization` ×14 (code map)
- `cpu-multi-modal-model-tests-n` ×4 (code map)
- `cpu-spec-decode-tests` (code map)

</details>

<details><summary>AMD mirrors: would skip (32)</summary>

- `basic-correctness` ×2
- `basic-correctness-cpu-offload`
- `basic-correctness-cumem`
- `basic-correctness-prefetch-offload`
- `basic-correctness-sleep-mode`
- `basic-models-tests-initialization`
- `basic-models-tests-other`
- `benchmarks-cli-test`
- `distributed-model-tests-2-gpus` ×3
- `entrypoints-integration-api-server` ×4
- `entrypoints-integration-api-server-generate`
- `entrypoints-integration-api-server-openai-chat_completion`
- `entrypoints-integration-api-server-openai-completion`
- `entrypoints-integration-llm`
- `entrypoints-integration-multimodal`
- `entrypoints-integration-pooling`
- `entrypoints-integration-responses-api`
- `entrypoints-integration-speech_to_text`
- `kernels-fla-ops-test-b200`
- `kernels-mhc-test-b200`
- `kernels-root-misc-test-b200`
- `language-models-tests-granite-l4-compatibility`
- `language-models-tests-hybrid` ×2
- `language-models-tests-standard`
- `multi-modal-models-standard-1-qwen2`
- `multi-modal-models-standard-2-qwen3-gemma`
- `multi-modal-models-standard-3-llava-qwen2-vl`
- `multi-modal-models-standard-4-other-whisper`
- `multi-modal-processor` ×4
- `platform-tests`
- `pytorch-compilation-passes-unit-tests`
- `rl-entrypoints-tests`

</details>

<details><summary>AMD mirrors: would add (0)</summary>

none

</details>

<sub>2 changed files · base `e572e02d21` · head `a14d7f0c2b` · Python record: build 93039 at `1e5d0ea888` · kernel record: table [`43b4aae`](https://github.com/aarnphm/aarnphm.github.io/commit/43b4aaea3e) (build 93244), map [`43b4aae`](https://github.com/aarnphm/aarnphm.github.io/commit/43b4aaea3e) · not counted: 10 build steps, 5 A100 steps the generator no longer emits, 10 optional steps the selector would also run</sub>

**elvircrn** (2026-10-07T21:07:12Z):

@mgoin Performance motivation should be covered by [this](https://github.com/vllm-project/vllm/pull/50656).

**mgoin** (2026-10-08T14:32:07Z):

/ci run

**github-actions** (2026-10-08T14:32:28Z):

✅ Triggered [Buildkite CI #93610](https://buildkite.com/vllm/ci/builds/93610) for commit `dda0106becc0`.

