Data checked 2026-08-13. Data may be stale - beyond the review cycle.
Review cycles are documented on the Methodology page. Methodology
torch2trt
Software Tool Tier A edge_ai MIT
An easy to use PyTorch to TensorRT converter
Engineering Snapshot
Use cases & tasks 5
- Best suited for
-
- Edge AI deployment on Jetson Nano/TX2/Xavier
- PyTorch-to-TensorRT model conversion workflows
- Low-latency classification/inference on embedded GPUs
- Primary tasks
-
- Convert PyTorch classification models to TensorRT
- Deploy PyTorch models on Jetson hardware for real-time inference
Stack & ecosystem
- Resource type
- Software Tool
- Ecosystem
- classification · inference · jetson-nano · jetson-tx2 · jetson-xavier · pytorch
License & compliance
- License
-
MIT(Inferred)
Lifecycle & freshness Show
- Maintenance
- Active
- Latest version
- Not recorded
- Last activity
- 2026-08-07
- Last checked
- 2026-08-13
- Verification
- Official confirmed
What It Solves
Converting PyTorch models to TensorRT engines for optimized inference on NVIDIA Jetson edge devices.
Primary use cases
- Convert PyTorch classification models to TensorRT
- Deploy PyTorch models on Jetson hardware for real-time inference
Secondary use cases
- Benchmark TensorRT vs PyTorch performance on Jetson
- Integrate into edge ML model optimization pipelines
When to Use
Consider when
- Targeting NVIDIA Jetson platforms (Nano, TX2, Xavier)
- Need MIT-licensed conversion tool
- Require active maintenance (last activity 2026-08-07)
Verify before adopting
- TensorRT version compatibility with target JetPack
- Operator coverage for specific model architectures
- Dynamic shape support limitations
Start Here
Adoption Checklist
- Needs verification TensorRT version compatibility with target JetPack
- Needs verification Operator coverage for specific model architectures
- Needs verification Dynamic shape support limitations
Each check stays "needs verification" until an official source confirms it; unconfirmed items are never marked verified.
Known Limitations & Unknowns
Not publicly verified
- Full operator support matrix
- Performance on latest JetPack releases
- Dynamic input shape handling
Alternatives & Related Tools
Related tools
- tensorrt — Integrates with
- jetson-inference — compatible with
How is it used?
Start from the recorded entry points below, then validate against the technical checklist.
- Official site https://github.com/NVIDIA-AI-IOT/torch2trt
- Repository https://github.com/NVIDIA-AI-IOT/torch2trt
- Documentation https://github.com/NVIDIA-AI-IOT/torch2trt
Technical checklist
- OK License identified Recorded: MIT
- OK Maintenance signal Active
- OK Verification status Official confirmed
- OK Source evidence attached 1 source record(s)
- NEEDS REVIEW Latest version recorded Not recorded
Official Links
Metadata & Governance
| License | MIT — Inferred |
|---|---|
| Commercial model | Unknown |
| Maintenance status | Active |
| Verification status | Official confirmed — Confirmed via the official repository API responses in SourceRefs below. |
| Latest version | Not recorded |
| Latest release | Not recorded |
| Last activity | 2026-08-07 |
| Last checked | 2026-08-13 |
| First seen | Not recorded |
Dataset facts
Facts above come from the official dataset card only; unconfirmed fields stay unknown.
Related Resources & Dependencies
- tensorrt — integrates with (confirmed)
- jetson-inference — compatible with (verified)
Recent Activity
New resource: NVIDIA-AI-IOT/torch2trt
New repository resource added by the sprint promote pipeline.
Related Knowledge
Guides
- Deploying AI Inference on UAV Edge Compute — A repeatable workflow for choosing and validating an inference runtime for an onboard UAV computer.
Collections
- Edge AI Perception Starter — A starting stack for prototyping aerial perception models before edge deployment work.