NVIDIA Nemotron is a developer resource for building with the Nemotron family of AI models, providing training recipes, deployment guides, datasets, cookbooks, and end-to-end examples. It is intended for developers and researchers training, customizing, deploying, or applying Nemotron models to agentic AI use cases. Its catalogue add-ons include a Claude Code plugin that guides users through Nemotron customization steps.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add NVIDIA-NeMo/Nemotron --skill airgapgit clone --depth 1 https://github.com/NVIDIA-NeMo/NemotronWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia-nemo/nemotron/airgap)<a href="https://agentmods.dev/skills/nvidia-nemo/nemotron/airgap"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/nemotron/airgap.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00089 | $0.01126 |
| Opus 5 | $0.00044 | $0.00563 |
| Sonnet 5 | $0.00018 | $0.00225 |
| Haiku 4.5 | $0.00009 | $0.00113 |
Grade A, and why
nemotron-customizer-airgap scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nemotron Customizer Airgap
Use this skill to help an agent produce a connected-machine airgap bundle and then submit Nemotron Customizer steps from the airgapped side. Keep it grounded in the checked-in runner and manifests; do not invent a parallel packaging flow.
Read First
deploy/nemotron-customizer/airgap/README.mdfor the operator flow.deploy/nemotron-customizer/airgap/airgap.yamlfor the current image map.deploy/nemotron-customizer/airgap/runner.pywhen changing behavior.tests/deploy/test_airgap_runner.pybefore editing runner logic.deploy/nemotron-customizer/airgap/configs/for runtime overlay configs.
For selected steps, inspect the catalog through the CLI:
uv run nemotron steps show <step_id> --json
Workflow
-
Establish the side of the workflow:
- Connected machine: validate, build, save image tarballs.
- Airgapped side: load images, set env profiles, run selected steps.
-
Gather the minimum inputs:
- Target steps and config names, for example
sft/megatron_bridge:tiny. - Target architecture or Docker platform, for example
linux/amd64. - Available base images and whether the connected machine can pull them.
- Airgapped env profile name, mounts, model/data/checkpoint locations.
- Whether destructive or expensive actions such as
--execute, Docker build, Docker volume cleanup, or state-file removal are explicitly allowed.
- Target steps and config names, for example
-
Plan with the runner first:
uv run python deploy/nemotron-customizer/airgap/runner.py \
--config deploy/nemotron-customizer/airgap/airgap.yaml
Use --target <step_id>:<config> for one-off selections without editing YAML.
The runner expands dependencies from dependencies, validates selected step
files/configs, groups execution images, and prints selected execution images.
- Edit
airgap.yamlonly where the runner expects configuration:workflow.stagesor CLI--targetfor selected customer steps.dependenciesfor explicit upstream Nemotron Customizer step outputs.step_execution_imagesfor step-to-image mapping.execution_imagesfor base image, tag, tar, platform, and import probes.launcher_imagefor the launcher container.
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- .gitignore 91 B
- airgap.yaml 4.0 KB
- configs/sft_megatron_bridge_default.yaml 449 B
- configs/sft_megatron_bridge_tiny.yaml 443 B
- Dockerfile.execution 1.7 KB
- Dockerfile.execution.dockerignore 373 B
- Dockerfile.launcher 711 B
- Dockerfile.launcher.dockerignore 349 B
- README.md 5.7 KB
- runner.py 45 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 116 lines · 89 tokens per session scan A d359ab5e9490
nemotron-customizer-airgap is a skill published in the GitHub repository NVIDIA-NeMo/Nemotron (2,023 stars, last pushed 4d ago), licensed Apache-2.0. It adds 89 tokens to every session and 1,126 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
tinker-fine-tuning
Provides guidance for fine-tuning LLMs using the Tinker cloud training API from Thinking Machines Lab. Use when running supervised fine-tuning, reinforcement learning (GRPO/PPO), or LoRA training on cloud GPUs via Tinker's managed infrastructure instead of local compute.
nvcf-self-managed-cli
Install, operate, and tear down self-hosted NVIDIA Cloud Functions (NVCF) deployments with nvcf-cli. Use for control-plane or compute-plane install, status checks, cluster registration, function deploy/invoke, task create/list/cancel/delete, API keys, admin tokens, JWKS rotation, failed-install diagnosis, and…
nvca-self-managed-install
Install or validate the NVCA Operator chart against a self-managed NVCF control plane from the native monorepo. Use when the control plane comes from deploy/stacks/self-managed and NVCA must be installed with stack-derived image repository settings.
nvcf-explore-stack
Navigate and explain the NVCF self-hosted stack inside the monorepo. Maps helmfile releases to their charts, image-source subtrees, helm hooks, namespaces, and needs: dependency chains. Reads deploy/stacks/self-managed/helmfile.d/.yaml.gotmpl and deploy/stacks/nvcf-compute-plane/helmfile.d/.yaml.gotmpl as the source…
fine-tuning-with-trl
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
tensorrt-llm
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.