Skip to main content

> ML_LIBRARIES_CATALOG_v1.0

Qualified ML Libraries

Every library is independently qualified with primary citations, supported version ranges, hardware accelerator separation, and real-world failure patterns.

Showing 18 of 161 Qualified Libraries & Tools (Page 8 of 9)Accelerator Mappings & Failure Modes
Observability & Drift
1.2.6Apache-2.0

DeepEval

Confident AI

The open-source pytest-native unit testing framework for LLM and RAG applications.

Training:
Not Supported (Inference Only)
Inference:
CPU
#text
Observability & Drift
0.90.1MIT

promptfoo

promptfoo

Developer-first CLI and CI/CD testing tool for prompt engineering, red teaming, and LLM evaluation.

Training:
Not Supported (Inference Only)
Inference:
CPU
#text
Observability & Drift
1.2.0MIT

TruLens

Truera / Snowflake

Open-source LLM evaluation and tracking library famous for formulating the RAG Triad.

Training:
Not Supported (Inference Only)
Inference:
CPU
#text
lifecycle-tracking
2.16.2Apache-2.0

MLflow

Databricks / Linux Foundation AI & Data

The industry-standard open-source platform for machine learning lifecycle management and model registries.

Training:
Not Supported (Inference Only)
Inference:
CPUCUDA
#tabular#text#image#audio
lifecycle-tracking
0.18.1MIT (Client SDK) / Commercial (Backend Server)

Weights & Biases

Weights & Biases Inc.

The premier deep learning experiment tracking and artifact lineage platform.

Training:
Not Supported (Inference Only)
Inference:
#tabular#text#image#audio#video
lifecycle-tracking
3.55.2Apache-2.0

DVC

Iterative.ai

Git-native data version control and reproducible ML pipeline management.

Training:
Not Supported (Inference Only)
Inference:
#tabular#image#audio#video#text
lifecycle-tracking
0.40.1Apache-2.0

Feast

Linux Foundation AI & Data / Feast Community

The leading open-source feature store for production machine learning and real-time inference.

Training:
Not Supported (Inference Only)
Inference:
CPU
#tabular
Inference Serving
1.3.6Apache-2.0

BentoML

BentoML

High-performance model serving framework with adaptive micro-batching and containerization.

Training:
Not Supported (Inference Only)
Inference:
CPUCUDAROCMMPS
#tabular#text#image#audio#video
Inference Serving
0.13.1Apache-2.0

KServe

Linux Foundation AI & Data / KServe Community

Standardized serverless, cloud-native model inference platform on Kubernetes.

Training:
Not Supported (Inference Only)
Inference:
CPUCUDAROCM
#tabular#text#image#audio
orchestration
1.9.0Apache-2.0

Kubeflow

Cloud Native Computing Foundation (CNCF)

The Cloud Native Computing Foundation (CNCF) platform for end-to-end machine learning on Kubernetes.

Training:
CPUCUDAROCMXPUTPU
Inference:
CPUCUDAROCM
#tabular#text#image#audio#video
orchestration
2.12.20Apache-2.0

Metaflow

Outerbounds / Netflix

Netflix's human-centric Python framework for building and scaling real-world data science workflows.

Training:
CPUCUDA
Inference:
CPUCUDA
#tabular#text#image#audio#video
orchestration
0.66.0Apache-2.0

ZenML

ZenML GmbH

Extensible, tool-agnostic MLOps framework that decouples pipeline code from cloud infrastructure.

Training:
CPUCUDA
Inference:
CPUCUDA
#tabular#text#image#audio#video
orchestration
1.13.0Apache-2.0

Flyte

Linux Foundation AI & Data / Union.ai / Lyft

Enterprise-grade, strongly-typed workflow automation and orchestration platform on Kubernetes.

Training:
CPUCUDAROCMXPUTPU
Inference:
CPUCUDAROCM
#tabular#text#image#audio#video
lifecycle-tracking
1.16.2Apache-2.0

ClearML

Allegro AI / ClearML Community

Complete open-source MLOps suite for experiment tracking, remote execution, and data management.

Training:
CPUCUDA
Inference:
CPUCUDA
#tabular#text#image#audio#video
client-edge-embedded
4.21.0Apache-2.0

TensorFlow.js

Google / TensorFlow Team

Google's machine learning library for training and deploying models in JavaScript, WebGL, and WebGPU.

Training:
CPUCUDA
Inference:
CPUCUDAWEBGPUWASM
#image#text#tabular#audio
client-edge-embedded
3.0.0Apache-2.0

Transformers.js

Hugging Face

Hugging Face's library for running state-of-the-art transformer models in JavaScript and WebGPU.

Training:
Not Supported (Inference Only)
Inference:
CPUWEBGPUWASM
#text#image#audio
client-edge-embedded
1.0.1MIT

ml5.js

ITP NYU / ml5.js Community

User-friendly, beginner-accessible machine learning library for the web and creative coding.

Training:
CPU
Inference:
CPUWEBGPUWASM
#image#video#audio#tabular
client-edge-embedded
20240410BSD-3-Clause

NCNN

Tencent

Tencent's ultra-optimized mobile neural network forward pass framework with ARM NEON assembly.

Training:
Not Supported (Inference Only)
Inference:
CPUWASM
#image#video