compare
LMDeploy vs TensorRT-LLM
The same facts for both, read from GitHub every night, and the relation a person reviewed.
| Fact | LMDeploy | TensorRT-LLM |
|---|---|---|
| Language | Python | Python |
| Licence | Apache-2.0 | Other |
| Stars | 8.1k | 15k |
| Latest | v0.18.0 | v1.2.1 |
| Last push | 2026-09-28 | 2026-10-06 |
| Release cadence | about 23 days between releases | too few releases |
| Active contributors | 28 commit authors on the default branch in the last 90 days | 125+ commit authors on the default branch in the last 90 days |
| Flags | none | none |
How they relate
Both replace ChatGPT. Alternatives to ChatGPT →
PartialLMDeployServes open-weight models behind an OpenAI-compatible API, with quantisation and batching.
PartialTensorRT-LLMServes open-weight models on NVIDIA GPUs behind an OpenAI-compatible API.
LMDeploy
TensorRT-LLM
- v1.3.0rc292026-09-29Model Supportpre-release
- v1.3.0rc282026-09-23Known Issuespre-release
- v1.3.0rc272026-09-18Known Issuespre-release
- v1.3.0rc262026-09-09Known Issuespre-release
- v1.3.0rc252026-08-31Annoucementspre-release
LMDeploy
✓ signed The latest release, v0.18.0, carries a signature GitHub verified.
Loading the security report
TensorRT-LLM
✓ signed The latest release, v1.2.1, carries a signature GitHub verified.
Loading the security report