cd /news/artificial-intelligence/vllm-merges-embedding-loader-fix-aft… · home › topics › artificial-intelligence › article
[ARTICLE · art-170482] src=runtimewire.com ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

vLLM merges embedding loader fix after 4.2 MB test allocated 3 GiB

A vLLM pull request merged into main on November 16th, 2026 moves the archive size check ahead of torch.load after contributor Leizhen Zhang showed a 4,175,968-byte base64 embedding payload allocated 3,072 MiB in 'safe_load_prompt_embeds' before the shape check rejected it. A second variant using a one-row view passed the 2 GiB check while retaining the full 3 GiB storage. The fix sums the declared uncompressed size inside the archive and rejects oversized payloads before loading begins.

read1 min views1 publishedNov 16, 2026
vLLM merges embedding loader fix after 4.2 MB test allocated 3 GiB
Image: Runtimewire (auto-discovered)

In a vLLM contributor's test, a 4,175,968-byte reproducer allocated 3,072 MiB before the shape check rejected it, and a one-row view passed with 3 GiB of storage attached. The fix merged into main on November 16th, 2026 checks archive size before torch.load.

        By [RuntimeWire Staff](https://runtimewire.com/author/runtimewire-staff)
        · Published 

Contributor Leizhen Zhang tested vLLM, an open-source inference and serving engine, with a 4,175,968-byte base64 embedding payload that allocated 3,072 MiB in 'safe_load_prompt_embeds' before the shape check rejected it. A second version using a one-row view passed the 2 GiB check while keeping the full 3 GiB storage attached. The merged pull request moves the check to the front of the path: the fix on vLLM's main branch sums the declared uncompressed size inside the archive and rejects oversized payloads before 'torch.load' runs.…

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @vllm 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/vllm-merges-embeddin…] indexed:0 read:1min 2026-11-16 · —