gpu-selection

gpu-selection is a skill for Claude Code, Codex from nebius/nebius-physical-ai. It costs 28 tokens per session (1,150 once invoked), scanned A, original, Apache-2.0.

A guide for selecting suitable GPUs, the computer processors used for demanding graphics and machine-learning work, in NPA workbench workflows.

In plain words
What is it for?
It is for choosing GPU types and counts for training, rendering, inference, command-line tools, SDK settings, and workflow YAML files.
Why use it?
It helps avoid assigning work to hardware that lacks the needed capabilities or cannot be scheduled.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Install

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.

agentmods
npx agentmods add skills/nebius/nebius-physical-ai/gpu-selection
Any agent
npx skills add nebius/nebius-physical-ai --skill gpu-selection
Clone the repo
git clone --depth 1 https://github.com/nebius/nebius-physical-ai

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for gpu-selection

README.md
[![agentmods](https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/gpu-selection.svg)](https://agentmods.dev/skills/nebius/nebius-physical-ai/gpu-selection)
Your own site
<a href="https://agentmods.dev/skills/nebius/nebius-physical-ai/gpu-selection"><img src="https://agentmods.dev/badge/skills/nebius/nebius-physical-ai/gpu-selection.svg" alt="Measured on agentmods" height="20"></a>
Per session 28 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,150 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00028 $0.01150
Opus 5 $0.00014 $0.00575
Sonnet 5 $0.00006 $0.00230
Haiku 4.5 $0.00003 $0.00115

Measured 6d ago against content hash 3e3415c25152, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

gpu-selection 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 6d 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.

skills/atomic/gpu-selection/SKILL.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GPU Selection

When To Use

Use this skill when a task asks which GPU family to use, changes workflow resources, updates image routing, or reviews render/training placement.

Procedure

  1. Identify whether the workload needs RT cores, tensor throughput, multi-GPU scaling, or only CPU resources.
  2. Check the tool-specific skill for hard constraints.
  3. Encode the choice in CLI flags, SDK config, or workflow YAML env/resources.
  4. Keep image variants aligned with GPU selection.
  5. For direct-Kubernetes Jobs, discover nvidia.com/gpu.product labels and construct an ordered, compatible candidate list. Move to the next product only for concrete scheduler evidence (Unschedulable, insufficient GPU resource, or no matching product/affinity); runtime, pull, credential, checkpoint, and application failures are not placement failures.

Three-Tier Contract

  • CLI: commands expose GPU choices through flags such as --gpu-type, --gpu-preset, --runtime, or tool-specific image variant options.
  • SDK: runtime config and request builders should carry GPU type/count rather than deriving it from private environment names.
  • YAML: workflow resources and env vars must express the GPU target explicitly enough for reviewers to validate routing.

Current Defaults

  • H100: general training, CLIP embedding, detection, MJLab, Cosmos inference, LeRobot training smoke, and non-render throughput.
  • L40S: Isaac Lab and SONIC render validation on VM hosts.
  • RTX PRO 6000 Blackwell: Isaac Lab and SONIC render validation on Kubernetes with mounted NVIDIA GPU Operator drivers.
  • B200 / B300: headless, state-based training and inference only.
  • CPU: Retargeting and many dataset curation/import steps.

Blackwell Is Two Different Targets

"Blackwell" spans two CUDA majors, and their binaries are mutually incompatible:

GPU Compute capability SM Nebius platform
RTX PRO 6000 Blackwell 12.0 sm_120 gpu-rtx6000
B200 10.0 sm_100 gpu-b200-sxm (us-central1)
B300 (Blackwell Ultra) 10.3 sm_103 gpu-b300-sxm

Read the full file on GitHub · 100 lines

Changes

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.

  1. 6d ago First seen · 100 lines · 28 tokens per session scan A 3e3415c25152

Subscribe to this mod's changes

gpu-selection is a skill published in the GitHub repository nebius/nebius-physical-ai (27 stars, last pushed today), licensed Apache-2.0. It adds 28 tokens to every session and 1,150 once invoked, about $0.0001 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.

Related

Other skills, from other repositories

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

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.

google/skills · 60 tokens

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.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens