---
cited_by:
  - '[[thoughts/papers/qiu2026designqwen38nextarchitectureevaluation#13-ple--n-gram-embedding-engram-style|qwen4next §1.3]]'
  - '[[thoughts/inference/qwen3.8-serving#4-pr-reading-order-vllm-main-git-show-sha|qwen4next §4]]'
  - '[[thoughts/inference/qwen3.8-serving#5-sharp-edges-bugs-that-expose-invariants|qwen4next §5]]'
date: '2026-11-11'
kind: pr
local:
  - file:///Users/aarnphm/workspace/vllm/vllm/triton_utils/importing.py
merged: '2026-10-06'
number: 60142
source:
  - https://github.com/vllm-project/vllm/blob/b0b23a48d466712d18d390e73b62b169842f65e2/vllm/models/qwen4_exp/common/ngram_embedding.py
  - https://github.com/vllm-project/vllm/blob/b0b23a48d466712d18d390e73b62b169842f65e2/vllm/triton_utils/importing.py
state: merged
tags:
  - vllm
  - vllm-pr
  - new-architectures
title: '#60142 [Bugfix] Fix Qwen4Exp FP8 PLE pinned lookup kernel failing to compile below SM89'
topic: new-architectures
url: https://github.com/vllm-project/vllm/pull/60142
created: '2026-11-11'
modified: '2026-11-11'
published: '2026-11-11'
pageLayout: default
slug: hinterland/papers/maps/prs/60142
permalink: https://aarnphm.xyz/hinterland/papers/maps/prs/60142.md
generator:
  quartz: v4.6.0
  hostedProvider: Cloudflare
  baseUrl: aarnphm.xyz
full: https://aarnphm.xyz/llms-full.txt
---
Fixes [#59868](https://github.com/aarnphm/aarnphm.github.io/issues/59868). On GPUs below SM89 (the report uses 4 × RTX 3090), Qwen4Exp with an FP8 E4M3FN PLE table and `--engram-config '{"cpu_offload": true}'` crashes in `profile_run`. The crash is a Triton `CompilationError` from… · full dump: [[thoughts/inference/vllm-prs/60142|#60142]]

