TensorRT-LLM
NVIDIAgithub.com/NVIDIA/TensorRT-LLM
TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate
Updated 54 min ago · first seen 11 Sept 2026
framework_01M294ES1YRZVZWYQP5W3K09DW
- Version
- 1.3.0rc26
- T2 · 3 h ago
- Released
- 9 Sept 2026
- T2 · 3 h ago
- Stars
- 14,597
- T2 · 54 min ago
Specification
- Repository
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Author
- NVIDIA Corporation
Source:PyPI public package metadataT2observed 57 min agomedium
- Created at
- 16 Aug 2023
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Default branch
- main
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Homepage
- https://nvidia.github.io/TensorRT-LLM
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Kind
- runtime
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Latest release
- 9 Sept 2026
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Latest release tag
- v1.3.0rc26
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Latest version
- 1.3.0rc26
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- License name
- Other
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Forks
- 2,736
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 54 min agomedium
- Releases
- 89
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 54 min agomedium
- Stars
- 14,597
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 54 min agomedium
- Tags
- 99
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 54 min agomedium
- Watchers
- 122
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 54 min agomedium
- PyPI
- tensorrt-llm
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Pypi license
- Apache License 2.0
Source:PyPI public package metadataT2observed 57 min agomedium
- Pypi release at
- 20 Apr 2026
Source:PyPI public package metadataT2observed 57 min agomedium
- Pypi url
- https://pypi.org/project/tensorrt-llm/
Source:PyPI public package metadataT2observed 57 min agomedium
- Pypi version
- 1.2.1
Source:PyPI public package metadataT2observed 57 min agomedium
- Python classifiers
- Programming Language :: Python :: 3.10, Programming Language :: Python :: 3.12
Source:PyPI public package metadataT2observed 57 min agomedium
- Readme url
- https://raw.githubusercontent.com/NVIDIA/TensorRT-LLM/HEAD/README.md
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Releases url
- https://github.com/NVIDIA/TensorRT-LLM/releases
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
- Requires python
- <4,>=3.10
Source:PyPI public package metadataT2observed 57 min agomedium
- Topics
- blackwell, cuda, llm-serving, moe, pytorch
Source:GitHub public repositories (HTML, releases.atom, raw files)T2observed 3 h agomedium
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
26
Source tiers
T226
Freshest observation
54 min ago
Conflicts
None
- Developed by
- NVIDIA
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Requires python
Requires pythonrequires_python1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| <4,>=3.10 | → current | current | PyPI public package metadataT2 | medium | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
NVIDIA/TensorRT-LLM released v1.3.0rc22.post1: [None][chore] Bump version to 1.3.0rc22.post1
githubNVIDIA/TensorRT-LLM released v1.3.0rc23.post1: [None][ci] Recover necessary CI stages for release (#17895)
github
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| GitHub public repositories (HTML, releases.atom, raw files) | github.com/NVIDIA/TensorRT-LLM/releases.atom | feed | T2· Quality secondary | 54 min ago | 1 |
| GitHub public repositories (HTML, releases.atom, raw files) | raw.githubusercontent.com/NVIDIA/TensorRT-LLM/HEAD/README.md | readme | T2· Quality secondary | 54 min ago | 1 |
| GitHub public repositories (HTML, releases.atom, raw files) | github.com/NVIDIA/TensorRT-LLM | repository | T2· Quality secondary | 54 min ago | 3 |
| PyPI public package metadata | pypi.org/pypi/tensorrt-llm/json | package | T2· Quality secondary | 57 min ago | 1 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.