AI Potluck
Model components / Inference code

vLLM

vLLM (UC Berkeley / vLLM team)

vLLM is a serving engine for large language models built on PagedAttention, which manages the KV cache as virtual memory pages to reduce memory waste during batched inference. It supports continuous batching, tensor and pipeline parallelism, speculative decoding, and quantization, exposing an OpenAI-compatible API server for self-hosted deployment. It originated in the Sky Computing Lab at UC Berkeley and is community-governed under vllm-project.

Verified 2026-08-09 via GitHub and the LICENSE body.

Openness

5 high confidence
5.0
license
Apache-2.0(OSI)
source
public
core-gated
ungated
governance
vLLM-project(community,PyTorch-ecosystem adjacent)

Apache-2.0 across the repo, re-confirmed today against the LICENSE body. The published tree is the full engine with no commercial edition or enterprise-licensed directory, so nothing is gated out of the source. Governance sits with the vLLM project community, adjacent to the PyTorch ecosystem.

  • https://github.com/vllm-project/vllm recorded 2026-08-09

    Public repository vllm-project/vllm, described as "A high-throughput and memory-efficient inference and serving engine for LLMs"; full source tree with no commercial edition or enterprise-licensed directory - the page's only "Enterprise"/"Pricing" strings are GitHub's own site navigation; README states it was originally developed in the Sky Computing Lab at UC Berkeley and is now maintained by a diverse community.

  • https://raw.githubusercontent.com/vllm-project/vllm/main/LICENSE recorded 2026-08-09

    "Apache License" "Version 2.0, January 2004" full text, unmodified.

Adoption

4 high confidence
4.0

PyPI downloads of 5.29M in the trailing 30 days band at level 4 (1M-10M); 88,585 GitHub stars corroborate but are not the primary basis.

Capability

5 high confidence
5.0

vLLM is the serving stack behind AMD's MLPerf Inference v6.0 submission on Instinct MI355X, covering gpt-oss-120b, Llama 2 70B and a multi-node Llama 3.1 405B run reported at 97-98% scaling efficiency, and the project reports support for 200+ model architectures on Hugging Face. It is the capability anchor for this category, the engine the others here are placed relative to.

Verified 2026-08-09