TensorRT-LLM

Frameworks

NVIDIA's inference optimization toolkit for LLMs

NVIDIA's optimized inference for maximum performance

Metrics

github stars9,500

radar statustrial

framework typeinference

weekly downloads180,000

adoption80

ecosystem86

performance95

performance35% weight

Execution speed and resource efficiency

Source: Benchmark comparisons, throughput measurements

adoption35% weight

Community size and industry usage

Source: GitHub stars, PyPI downloads, job postings

ecosystem30% weight

Integrations, plugins, and extension availability

Source: Integration count, ThoughtWorks Radar status

Last updated: December 24, 2025