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 nemo-megatron-bridgegit 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/nemo-megatron-bridge)<a href="https://agentmods.dev/skills/nvidia-nemo/nemotron/nemo-megatron-bridge"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/nemotron/nemo-megatron-bridge/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nvidia-nemo/nemotron/nemo-megatron-bridge"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/nemotron/nemo-megatron-bridge.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00113 | $0.01888 |
| Opus 5 | $0.00056 | $0.00944 |
| Sonnet 5 | $0.00023 | $0.00378 |
| Haiku 4.5 | $0.00011 | $0.00189 |
Grade A, and why
nemotron-3-ultra-text2sql-lora 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 10d 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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Nemotron-3 Ultra Text2SQL LoRA — runbook for a coding agent
This skill helps you run the cookbook in this directory (mbridge_lora_cookbook.ipynb) on the
user's behalf. The notebook is generic and ships with placeholders; your job is to gather the
user's environment details, fill them in, launch the SLURM jobs, watch them, and report results.
What the tutorial does
Three steps, in order, each a SLURM job:
- Data prep — builds a BIRD Text2SQL
training.jsonlfrom both the no-reasoning and reasoning splits, formatted with Ultra's tokenizer/chat template. Short CPU job. - Convert — distributed import of the Hugging Face base checkpoint into Megatron-Bridge format. A multi-node GPU job (CPU import is not feasible for a 550B model).
- LoRA fine-tune — packed-sequence LoRA training on the prepared data; saves a LoRA adapter. A multi-node GPU job.
What you must understand before running
- Ultra is a 550B-total / A55B-active hybrid Mamba-Transformer MoE. It does not fit on one
node, so every heavy step is a multi-node SLURM job submitted with
sbatchand run in a container via Pyxis/enroot. Run everything from a cluster login node wheresbatch/squeue/sacctare available. - Scale. At the shipped parallel settings, both convert and train need 48 GPUs. Node count
is derived automatically as
48 / GPUS_PER_NODE(e.g. 12 nodes at 4 GPUs/node). The user's QOS must permit a job of that size — an interactive or small-node-capped QOS will not work. - Single config. Everything is driven by one file,
config.env, which the notebook's setup cell generates from the values you fill in. Every step and everyslurm/*.sbatchscript sources it. You can run the notebook cell, or writeconfig.envdirectly with the same keys. - One output root.
WORKSPACEis the single output root; everything generated lands under$WORKSPACE/{base, dataprep, trained, cache/hf, logs}. The base checkpoint (HF_MODEL_PATH) is the only separate, read-only path. - The rhythm per step: a launch cell submits the job, a re-runnable check cell shows
status (
sacct/squeue), and a sanity cell confirms the expected output exists before you move on. Follow this loop; don't skip the sanity check.
What ships with it
11 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.
- base_sft_dataset.py 4.6 KB runs code
- dataprep.py 2.3 KB runs code
- dataset_bird_reasoning.py 1.5 KB runs code
- dataset_bird.py 1.2 KB runs code
- mbridge_lora_cookbook.ipynb 15 KB
- README.md 3.5 KB
- slurm/convert.sbatch 2.2 KB
- slurm/dataprep.sbatch 1014 B
- slurm/train_lora.sbatch 1.8 KB
- TODO.md 2.4 KB
- train_lora.py 4.4 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.
- 10d ago First seen · 129 lines · 113 tokens per session scan A 55cee9430897
nemotron-3-ultra-text2sql-lora is a skill published in the GitHub repository NVIDIA-NeMo/Nemotron (2,034 stars, last pushed 3d ago), licensed Apache-2.0. It adds 113 tokens to every session and 1,888 once invoked, about $0.0006 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.
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.
huawei-cloud-cloudrobo-train
Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAINFROMSCRATCH) and finetune (MODELTUNING) tasks with FFT/SFT/LORA/QLORA/DEEPSPEED methods; manage the full task lifecycle (create/read/update/delete/stop/restart/resume/draft); save and resubmit draft…
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.
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.
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.