tensorrt-cu12-bindings
A high performance deep learning inference library
Decision gist · record as of 2026-08-14
Yes, if you have CUDA 12 and an NVIDIA GPU and need to deploy deep learning inference at scale. The package is actively maintained, has no known vulnerabilities, and is part of a mature inference platform. Verify licensing terms with NVIDIA for your use case, and confirm that CUDA 12 and compatible drivers are already installed before attempting installation.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires CUDA 12 runtime and compatible NVIDIA GPU drivers; only available for Python 3.8–3.14 on Linux x86_64 (manylinux_2_28) or Windows x86_64.
- Medium install friction: platform-specific wheels for Python 3.8–3.14 on Linux (manylinux_2_28) and Windows x86_64 only.
- No runtime dependencies, but requires CUDA 12 and compatible GPU drivers already present.
License · maintenance · safety
Proprietary (unclear) — Licensed as Proprietary with unclear treatment. Users should verify licensing terms with NVIDIA before deploying in production or commercial contexts.
last release 2026-07-30 (15 days) · last repo commit 2026-08-04 · 13,250 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 297,245 downloads/mo, #7,887 on PyPI
Alternatives
Verify before relying
pip install tensorrt-cu12-bindings
import tensorrt as trt
logger = trt.Logger(trt.Logger.WARNING)
builder = trt.Builder(logger)- Whether this package alone is sufficient for inference or if additional TensorRT components must be installed separately.
- Exact CUDA 12 minor version compatibility and minimum driver version required.
- Whether the package includes prebuilt TensorRT libraries or requires a separate TensorRT GA build installation.
- Supported precision levels and quantization capabilities beyond explicit quantization mentioned in the description.
What it is and what it does
tensorrt-cu12-bindings is a Python interface to NVIDIA's TensorRT inference engine, version 11.2.1.2, compiled for CUDA 12. It allows developers to load, optimize, and execute deep learning models on NVIDIA GPUs with minimal latency and maximum throughput. The package provides strongly-typed network APIs and explicit quantization support as part of TensorRT 11.X's redesigned architecture.
The bindings are distributed as platform-specific wheels for modern Python versions on Linux x86_64 and Windows x86_64. Installation is straightforward via pip, but requires CUDA 12 and compatible GPU drivers to be present on the system. There are no Python runtime dependencies; the package is a thin wrapper around compiled TensorRT libraries.
Use it for
- Optimize and deploy pre-trained deep learning models for low-latency inference on NVIDIA GPUs.
- Build quantized inference engines using explicit quantization APIs to reduce model size and improve throughput.
- Integrate custom CUDA kernels and operations via IPluginV3 plugin API for domain-specific inference acceleration.
- Benchmark and profile inference performance across different batch sizes and precision levels.
- Deploy inference services in production environments where model latency and throughput are critical.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you have CUDA 12 and an NVIDIA GPU and need to deploy deep learning inference at scale.
The package is actively maintained, has no known vulnerabilities, and is part of a mature inference platform. Verify licensing terms with NVIDIA for your use case, and confirm that CUDA 12 and compatible drivers are already installed before attempting installation.
Install
tensorrt-cu12-bindings on PyPI
Before you install
Medium install friction: platform-specific wheels for Python 3.8–3.14 on Linux (manylinux_2_28) and Windows x86_64 only. No runtime dependencies, but requires CUDA 12 and compatible GPU drivers already present. Active maintenance with recent releases.
Requires CUDA 12 runtime and compatible NVIDIA GPU drivers; only available for Python 3.8–3.14 on Linux x86_64 (manylinux_2_28) or Windows x86_64.
License in practice
Licensed as Proprietary with unclear treatment. Users should verify licensing terms with NVIDIA before deploying in production or commercial contexts.
Quickstart
pip install tensorrt-cu12-bindings
import tensorrt as trt
logger = trt.Logger(trt.Logger.WARNING)
builder = trt.Builder(logger)Verify before relying
- Whether this package alone is sufficient for inference or if additional TensorRT components must be installed separately.
- Exact CUDA 12 minor version compatibility and minimum driver version required.
- Whether the package includes prebuilt TensorRT libraries or requires a separate TensorRT GA build installation.
- Supported precision levels and quantization capabilities beyond explicit quantization mentioned in the description.
Package facts
| License | Proprietary unclear |
| Python support | Not specified |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | None |
| Maintenance | Actively maintained 15 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 297,245 / month, #7,887 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Intended Audience :: DevelopersLicense :: Other/Proprietary LicenseProgramming Language :: Python :: 3 |
Evidence: tensorrt_cu12_bindings-11.2.1.2-cp310-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp310-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp311-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp311-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp312-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp312-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp313-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp313-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp314-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp314-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp38-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp38-none-win_amd64.whl; tensorrt_cu12_bindings-11.2.1.2-cp39-none-manylinux_2_28_x86_64.whl; tensorrt_cu12_bindings-11.2.1.2-cp39-none-win_amd64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “tensorrt cuda 12 python bindings”
- tensorrt-cu12-bindingsProvides Python bindings for NVIDIA TensorRT 11.2.1.2 compiled for…
- tensorrt-cu12Provides Python bindings for NVIDIA TensorRT 11, enabling…
- tensorrt-cu12-libsProvides NVIDIA TensorRT libraries for CUDA 12 environments, enabling…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also tensorrt-cu12 · tensorrt-cu13-bindings · tensorrt-cu12-libs · tensorrt-cu13 · tensorrt-cu13-libs · tensorrt · dyNET38 · sit4onnx · transformer-engine-cu12 · nvidia-cublas-cu11