
Deploy Model Garden models to Agent Platform endpoints with cost checks and cleanup
- Open source
- 24.4K GitHub stars
- Cloud and infrastructure

Deploy Model Garden models to Agent Platform endpoints with cost checks and cleanup

Practitioner runbook for training and serving large models, from GPUs to NaN loss spikes

Guides agents writing Gradio code for ZeroGPU Spaces, covering @spaces.GPU limits

Lets an agent submit, monitor and cancel TAO GPU training jobs on Kubernetes clusters

Guides coding agents to write, run and debug GPU-accelerated cuML machine learning code

Helps an agent find matching Earth2Studio weather and climate models, data sources

Runs Megatron-LM and Megatron Bridge training and checks their loss curves match

Guides choosing and combining TP, DP, PP, CP and EP degrees for Megatron Bridge training

Runs iterative NeMo-RL experiments with git branches and TSV logs as a research ledger

Runs NVIDIA PAIDF Cosmos Predict to generate videos from a media sample list

Computes image embeddings from a parquet of file paths using CLIP, SigLIP or TAO

Gates TAO model launches behind preflight checks, image confirmation and a review step

Mines look-alike source images for VCN augmentation via DEFT embedding
Sourced AI and cloud pricing for FinOps: compare LLM costs, hosts and GPU break-even
Rent NVIDIA GPU pods, run commands on them and remove them, billed per second
Runs locallyLets coding assistants rent GPUs by the minute, run jobs and shut them down within budgets
Not checked yetRent EU cloud GPUs by the minute from an agent, with a hard hourly budget per launch

Look up AI model and agent pricing and estimate whether a model fits your GPU
Not checked yetChat with open-weight models running on the open c0mpute node network
Not checked yetShared GPU compute beta with an account API, task catalog and MCP access
Not checked yetLets a coding assistant rent a per-minute GPU, run a job and shut it down
Runs locally
Search measured local-LLM speeds on home GPU rigs and check what will run
Not checked yetDaily prices, stock and measured speeds for computers that run local AI models
Not checked yetLet an agent find, launch and monitor rented GPUs with cost checks
Not checked yetCheck whether an open LLM fits your GPU, with VRAM math and what fits instead
Not checked yetMCP server to rent A100 and H100 GPUs by the minute from your agent within a spending cap
Not checked yetRent GPUs at live prices and serve Hugging Face models, within a spend cap you set
Not checked yetQuery AI inference price indexes, model prices, speed and cost per task
Not checked yetOpen models ranked by momentum, with GPU or Mac fit verdicts and GPU rental prices
Not checked yetSearch Aguasto's published GPU and model research, plus AI infrastructure calculators
Not checked yetThe most popular MCP servers and agent skills for gpu on AI-Powered right now are Agent platform deploy, ML engineering, Huggingface zerogpu, Tao run on kubernetes and Cuml. Rankings combine upvotes, recent activity and, for open-source projects, GitHub stars.
Yes. Of the 51 MCP servers and agent skills for gpu listed, 35 are free or open source and 4 have a free plan. Use the Pricing filter to see only those.
AI-Powered lists 51 MCP servers and agent skills for gpu. New listings are added daily from the official MCP Registry, GitHub and maker submissions.