Filter by
(2)
IBM's Granite 3.0 large language model (LLM), optimized for local large language model operations
2y
10K+
1
IBM's Granite 3.0 large language model (LLM), optimized for local large language model operations
2y
10K+
8
IBM's Granite 3.0 large language model (LLM), optimized for local large language model operations
2y
10K+
4
IBM's Granite 3.0 large language model (LLM), optimized for local large language model operations
2y
10K+
1
IBM's Granite 3.1 large language model (LLM), optimized for local large language model operations
1y
10K+
9
A python-based machine learning framework, providing tensors, dynamic neural networks and strong GPU acceleration.
4h
500K+
7
Langfuse is an open-source LLM engineering platform for observability, prompt management, evaluations, and datasets.
10h
500K+
Background job processor for the Langfuse LLM engineering platform, handling evaluations, ingestion, batch exports, and data retention.
10h
100K+
LiteLLM is a universal LLM API gateway that provides OpenAI-compatible APIs for 100+ LLM providers including Bedrock, Azure, OpenAI, Anthropic, Vertex AI, and more.
10h
100K+
The Kubeflow Pipelines provides the UI of the Kubeflow Pipeline ecosystem.
10h
100K+
The Robot Operating System (ROS) is an open source project for building robot applications.
6d
10M+
726
MLflow is an open-source platform for managing the machine learning lifecycle: experiment tracking, reproducible runs, model packaging, and deployment. This image provides the mlflow CLI and server runtime packaged as a hardened container for secure production deployments.
10h
100K+
Downloads model artifacts from various storage backends (S3, GCS, Azure, etc.) before KServe model serving containers start.
4h
100K+
NVIDIA device plugin for Kubernetes that exposes GPUs to the Kubernetes scheduler, provides GPU health monitoring, and supports MIG, MPS, time-slicing, and CDI.
10h
100K+
4
The Kubeflow Pipelines API Server coordinates k8s workflows in the Kubeflow Pipeline ecosystem.
4h
100K+
A Kubernetes Custom Resource Definition for serving machine learning (ML) models on arbitrary frameworks.
10h
50K+
LiteLLM Database is the database-enabled LiteLLM proxy image with Prisma support for persistent PostgreSQL-backed keys, budgets, spend tracking, and gateway configuration.
4h
50K+
The Official Docker Image of OpenSearch (https://opensearch.org/)
8h
100M+
198
KServe Agent is a component that manages model serving workloads in KServe.
4h
50K+
NVIDIA Data Center GPU Manager exporter for Prometheus. Exposes GPU telemetry (utilization, memory, temperature, power, profiling) over an HTTP /metrics endpoint via the DCGM library.
10h
50K+
The routing component for KServe inference graphs, enabling complex model serving patterns with intelligent traffic routing.
4h
50K+
NVIDIA CUDA base image. The runtime variant ships the CUDA runtime libraries (cudart, cublas, cufft, curand, cusolver, cusparse, npp); the dev variant adds the nvcc compiler and development headers. Intended as a base image for GPU-accelerated workloads.
4h
10K+
Enables local model node management and caching.
10h
10K+
Enables local model caching and management.
10h
10K+
A flexible, high-performance serving system for machine learning models designed for production environments
10h
10K+
NVIDIA Federated Learning Application Runtime Environment for privacy-preserving distributed AI.
4h
10K+
A Model Context Protocol server that connects AI assistants to SmartBear testing and monitoring tools, including BugSnag, Reflect, Swagger, PactFlow, QMetry, QTM4J, Zephyr, and Collaborator.
4h
10K+
A high-performance vector database and similarity search engine for building the next generation of AI applications.
4h
10K+
The NVIDIA GPU Operator automates the management of the NVIDIA software components needed to provision GPUs in Kubernetes clusters. This image ships the operator controller, the nvidia-validator, the vectorAdd CUDA validation sample, and kubectl, together with the operand manifests the controller renders at runtime and the operator CRDs.
4h
10K+
3