LightLLM

license

Apache-2.0(OSI)

source

public(the published repository is the engine)

core-gated

ungated(no enterprise path in the repository root and no paid build of the engine in the README)

The LICENSE body is the stock Apache-2.0 text with no appended condition, read in full rather than taken from the API's label. The repository is public and unarchived and builds the engine itself. Its root tree carries no enterprise, ee or commercial directory and its README describes no licence-gated build, so the core reads as ungated. Confidence is medium because that is a repository-and-README read rather than a pricing-page read.

• https://raw.githubusercontent.com/ModelTC/LightLLM/main/LICENSE recorded 2026-08-31

  LICENSE body, read in full: the stock Apache-2.0 text with no appended condition.

• https://api.github.com/repos/ModelTC/LightLLM recorded 2026-08-31

  Repo metadata for ModelTC/LightLLM - archived false, 4,253 stars.

• https://raw.githubusercontent.com/ModelTC/LightLLM/main/README.md recorded 2026-08-31

  README, read for a paid tier, an enterprise edition or a licence-gated build beside the published source, and for whether any component is stated to be private.

4,253 GitHub stars, in the 1K-10K band of the stars scale. A `lightllm` package exists on PyPI but publishes no repository link, so it could not be confirmed to belong to this project and is not declared.

• https://api.github.com/repos/ModelTC/LightLLM recorded 2026-08-31

  Repo metadata - stargazers_count = 4,253 - for ModelTC/LightLLM.

Two bands below the vllm anchor and level with text-generation-inference, which records the comparable set - continuous batching, tensor parallelism, Flash Attention, quantization - at the same band. Placed there rather than higher because it composes those techniques rather than publishing a serving record of its own.

• https://raw.githubusercontent.com/ModelTC/LightLLM/main/README.md recorded 2026-08-31

  README, read for the engine's feature set: Python-based serving with a lightweight design and horizontal scalability, assembled from FasterTransformer, TGI, vLLM and FlashAttention techniques.