Skip to content

MCP servers and agent skills for nvidia

39 agent skills and 3 MCP servers for nvidia in AI models and LLM tooling, including Evaluate cluster, Tao run deft aoi cosmos3 and Tao run deft cr its mining, most popular first. 39 of the 42 are listed as free or open source. Each one is written up from its own site or README, with pricing, platforms and alternatives.

Showing 1–30 of 36

  1. Run a cloud-trial acceptance test on a GPU cluster and write a dated report

    • Open source
    • 19.7K GitHub stars
    • Cloud and infrastructure
  2. Runs an iterative DEFT loop that improves Cosmos3 bare OK/NG AOI inspection accuracy

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  3. Runs an iterative mining-based DEFT loop to improve Cosmos Reason traffic-video

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  4. Lets an agent submit, monitor and cancel TAO GPU training jobs on Kubernetes clusters

    • Open source
    • 4K GitHub stars
    • Developer tools
  5. Trains, evaluates, exports and runs DINO 2D object detectors via TAO in Docker

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  6. Runs a standard TAO train, evaluate and export sequence without AutoML loops

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  7. Checks progress of a detached KERMT training run: container state, logs and metrics

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  8. Builds synthetic datasets and generation pipelines with the Data Designer library

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  9. Guides an agent through adding a new training method to the NVIDIA FastGen codebase

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  10. Guides an agent through adding a new modality to NVIDIA FastGen across data, model

    • Open source
    • 4K GitHub stars
    • Design and media
  11. Builds or adapts NVIDIA FastGen experiment configs for a network, task and method

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  12. Guides an agent through generating and inspecting NVIDIA FastGen teacher and student

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  13. Guides an agent through running, resuming and debugging NVIDIA FastGen training jobs

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  14. Recommends an LLM serving runtime and memory flags for NVIDIA Jetson devices

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  15. Guides an agent through launching a vLLM or SGLang OpenAI-compatible server on Jetson

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  16. Adds EAGLE-3 or draft-model speculative decoding to a Jetson vLLM server

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  17. Guides an agent through building and changing NeMo AutoModel training and eval recipes

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  18. Runs Megatron-LM and Megatron Bridge training and checks their loss curves match

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  19. Configures CPU offloading in Megatron Bridge to cut GPU memory during training

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  20. Guides an agent through reducing peak GPU memory in Megatron Bridge training runs

    • Open source
    • 4K GitHub stars
    • Knowledge and memory
  21. Guidance for choosing FSDP or 3D parallelism when training MoE vision-language models

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  22. Guides Megatron-Bridge sequence packing and long-context training setup and validation

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  23. Guides an agent through improving NVIDIA Riva/NeMo speech recognition for a domain

    • Open source
    • 4K GitHub stars
    • AI models and LLM tooling
  24. Guides an agent through NVIDIA Nemotron embed and rerank retrieval recipes safely

    • Open source
    • 4K GitHub stars
    • Knowledge and memory