NVIDIA Model Optimizer is a library that changes deep-learning models through techniques such as quantization, pruning, distillation, neural architecture search, speculative decoding, and sparsity so they can run more efficiently. Developers use it to optimize Hugging Face, PyTorch, or ONNX models and export checkpoints for inference frameworks such as SGLang, TensorRT-LLM, TensorRT, or vLLM. The catalogue entries provide skills, instructions, and a plugin for using its workflows.
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/Model-Optimizer --skill speculative-decodinggit clone --depth 1 https://github.com/NVIDIA/Model-OptimizerWrote 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/model-optimizer/speculative-decoding)<a href="https://agentmods.dev/skills/nvidia/model-optimizer/speculative-decoding"><img src="https://agentmods.dev/badge/skills/nvidia/model-optimizer/speculative-decoding/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/model-optimizer/speculative-decoding"><img src="https://agentmods.dev/badge/skills/nvidia/model-optimizer/speculative-decoding.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00135 | $0.01308 |
| Opus 5.5 | $0.00054 | $0.00523 |
| Sonnet 5.5 | $0.00027 | $0.00262 |
| Haiku 4.5 | $0.00014 | $0.00131 |
Grade A, and why
speculative-decoding 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 2d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Speculative Decoding Draft-Model Training
Everything needed to take a target model from "no draft head" to "validated acceptance rate" lives in this directory. Work through the stages below in order for a new model; jump straight to a stage when you already know which one you need.
Two axes: stage and algorithm
The pipeline is the same shape for every draft-model algorithm — synthesize data, dump base-model hidden states, train the draft, benchmark acceptance rate. What changes between algorithms is which training script and recipe run, which knobs matter, and which failures are typical.
So this skill is split along those two axes, and you almost always read one file from each:
| Axis | Directory | What it holds |
|---|---|---|
| Stage | references/stages/ |
The procedure — algorithm-independent |
| Algorithm | references/algorithms/ |
The data sheet — scripts, recipe, knobs, thresholds, known failures |
Read the stage file for what to do, and the algorithm sheet for the values to
plug in. When a stage file says "see the algorithm sheet", it means the section of
references/algorithms/<algorithm>.md with the matching heading.
Where the code lives
The launcher scripts in tools/launcher/common/{eagle3,specdec}/ are thin wrappers around
examples/speculative_decoding/ — main.py (training), collect_hidden_states/ (dumps),
scripts/{export_hf_checkpoint,ar_validate,quantize_drafter}.py. Read a wrapper for the
flags it passes, that directory for what they do. Its README.md is also the non-launcher
path: training invoked directly, without Slurm.
Stages
| Stage | Reference | Use when |
|---|---|---|
| 1. Configure | references/stages/configure.md |
Adding a model that has no pipeline YAML yet |
| 2. Review logs | references/stages/review-logs.md |
A run finished (or died) and you want a pass/fail summary |
| 3. Triage | references/stages/triage.md |
A task failed and you need root cause plus a fix |
| 4. Validate | references/stages/validate.md |
All tasks passed and you need to confirm the acceptance rate gate |
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.
- references/algorithms/dflash.md 11 KB
- references/algorithms/dflash2.md 9.5 KB
- references/algorithms/domino.md 7.0 KB
- references/algorithms/dspark.md 11 KB
- references/algorithms/eagle3.md 6.2 KB
- references/algorithms/lilicorr.md 13 KB
- references/algorithms/README.md 2.5 KB
- references/stages/configure.md 2.2 KB
- references/stages/review-logs.md 2.9 KB
- references/stages/triage.md 6.2 KB
- references/stages/validate.md 4.9 KB
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.
- 2d ago Changed · +13 lines 2bb267acc9c9
- 1mo ago First seen · 93 lines · 135 tokens per session scan A 5792b323d876
speculative-decoding is a skill published in the GitHub repository NVIDIA/Model-Optimizer (5,398 stars, last pushed today), licensed Apache-2.0. It adds 135 tokens to every session and 1,308 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-09-18.
Other skills, from other repositories
agent-platform-model-registry
Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.
google-cloud-solution-agentic-analytics-spark-knowledge-catalog
Discovers requirements and designs an end-to-end governed agentic analytics solution using Knowledge Catalog and Managed Service for Apache Spark (Lightning Engine). Use when designing data science and analytics workflows across structured and unstructured distributed data (including in S3, Azure Blob, AlloyDB, and…
training-check
Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
finetuning
Fine-tune models on Microsoft Foundry using SFT (supervised), DPO (preference), or RFT (reinforcement with graders). Covers dataset preparation, training job submission, deployment, and evaluation. USE FOR: fine-tune, SFT, DPO, RFT, training data, grader, distillation, fine-tuned model, training job, large file…