Open-source desktop app to run, serve and fine-tune text, image, video and audio models entirely on your own machine.
GPU
Accelerated computing on GPUs, from CUDA workloads to rented inference and training capacity.
11 apps, 17 skills and 3 MCP servers tagged GPU.
Apps
Inference and training platform for open models: serverless per-token APIs, dedicated GPU deployments, and managed fine-tuning with SFT, DPO and RFT.
NVIDIA GPU cloud for training and inference: self-serve on-demand instances billed by the minute, 1-Click Clusters, and large private superclusters.
GPU cloud for AI developers: on-demand GPU pods, serverless GPU endpoints that scale to zero, and multi-node clusters, billed by the second.
Simulate how fast a local LLM runs on a given GPU or Mac before you buy the hardware.
Production inference for open and custom models — fast, autoscaling, and deployable into your own cloud.
Serverless GPUs from a Python decorator — deploy models and batch jobs with no containers or cluster to manage.
Inference, fine-tuning, and GPU clusters for open models — the full stack for teams building on open weights.
A generative media platform tuned for speed — image, video, and audio models served with very low latency.
Run thousands of open models with one API call, and deploy your own without touching Kubernetes.
Inference on custom LPU hardware, built for latency — open models served at speeds general-purpose GPUs struggle to match.
Skills
Diagnoses wrong gradients in differentiable NVIDIA Warp programs by measuring first — comparing autodiff against finite differences on a shrunk reproduction before proposing any fix.
Skill: Hugging Face Vision Trainer
by Hugging Face
Fine-tune object detection, image classification and SAM/SAM2 segmentation models on Hugging Face Jobs cloud GPUs, with dataset validation before you spend GPU minutes.
AMD's official skill for benchmarking vLLM/SGLang inference and driving GPU kernel optimization with Magpie and TraceLens trace analysis.
AMD's official skill for standing up a vLLM endpoint on Instinct MI300X/MI325X/MI350X/MI355X GPUs — detection, recipe lookup, launch and health check in one flow.
Skill: NVIDIA TAO RT-DETR
by NVIDIA
Trains, distils, quantises, evaluates, exports and runs inference for RT-DETR real-time object detection in NVIDIA TAO, routing training through AutoML hyperparameter search by default.
Skill: NVIDIA Holoscan SDK Setup
by NVIDIA
Inspects the host's hardware, OS, CUDA driver and existing tooling, recommends one Holoscan SDK install method with a reason, then hands off to the matching install skill.
Skill: NVIDIA NV-Reason-CXR
by NVIDIA
Runs NVIDIA's NV-Reason-CXR-3B chest X-ray reasoning model through a documented wrapper, locally on GPU or via the public Hugging Face Space, and returns the model's full reasoning trace as engineering evidence.
Skill: SageMaker HyperPod Cluster Debugger
by AWS Labs
Read-only diagnostics for cluster-wide SageMaker HyperPod failures on EKS or Slurm — CloudFormation errors, EFA health checks, lifecycle scripts, capacity, dangling nodes and autoscaler conflicts.
Skill: Jetson LLM Serve
by NVIDIA
Picks the right vLLM or SGLang runtime for your Jetson generation and JetPack version, then produces a working OpenAI-compatible serving command.
Skill: Portfolio Optimization with cuOpt
by NVIDIA
GPU-accelerated Mean-CVaR and Mean-Variance portfolio construction with NVIDIA cuOpt: scenario generation, variance-capped SOCP allocations, efficient frontiers, backtests and rebalancing.
Turns a plain-language weather question into a working Earth2Studio inference script — picks the AI forecast model, a compatible data source, an IO backend, and the step count.
NVIDIA's official skill for deploying and operating Nemotron Speech (Riva) NIMs — ASR, text-to-speech and translation, cloud-hosted or self-hosted on your own GPUs.
Skill: NVIDIA DALI Dynamic Mode
by NVIDIA
NVIDIA's official skill for DALI's imperative dynamic-mode API — write GPU data loading as ordinary Python, or migrate an existing pipeline-mode graph across.
Skill: NVIDIA CUDA-Q Onboarding
by NVIDIA
NVIDIA's official onboarding skill for CUDA-Q — installs the platform, writes your first quantum kernel, picks a GPU simulator and routes you to real QPU hardware.
Skill: NVIDIA cuDF Accelerated Computing
by NVIDIA
NVIDIA's own guidance for moving pandas workloads onto GPU DataFrames with cuDF and dask-cuDF — when it pays off, and how to keep results identical.
Skill: LLM Fine-Tuning with TRL
by Hugging Face
Fine-tune language and vision models with TRL or Unsloth — supervised, preference, and reinforcement training methods.
Skill: Hugging Face Spaces
by Hugging Face
Deploy and maintain applications on Hugging Face Spaces — SDK choice, GPU hardware, model loading, and debugging.
MCP servers
Replicate's official MCP server: search thousands of hosted models, read their schemas, and run predictions on image, video, audio and language models from inside an agent.
MCP: NVIDIA CUDA Docs
by NVIDIA
NVIDIA's official remote MCP server for CUDA: grounds a coding agent in first-party CUDA documentation and code samples instead of its training data.
MCP: Runpod MCP Server
by Runpod
Runpod's official MCP server for driving GPU infrastructure — create and manage Pods, Serverless endpoints, templates and network volumes from an AI client.
Related tags
Tags that appear alongside this one, ranked by how often.