# Gpu Sharing Advisor

> - Decide between MIG, MPS, and time-slicing for sharing NVIDIA GPUs across Kubernetes workloads, and produce the config to implement the choice. Use whenever the user asks about GPU sharing, fractional GPUs, partitioning a GPU, MIG profiles, Multi-Process Service, time-slicing, GPU oversubscription, virtual GPUs, "our GPUs sit idle", low GPU utilization, running many small models on one GPU, packing inference replicas, or letting multiple researchers or notebooks share an A100/H100/H200/L4. Also use when reviewing a time-slicing ConfigMap, a MIG strategy (single vs mixed), or nvidia.com/mig- resource requests. For scaling the shared workloads up and down use sibling gpu-autoscaling-engineer; for per-team GPU quotas use researcher-tenancy-provisioner; for the cost report that motivates sharing use gpu-cost-optimizer.

## Facts
- Page: https://tashan.sh/capability/skill-cloud-byte-consulting-gpu-sharing-advisor
- tashan id: skill:Cloud-Byte-Consulting/gpu-sharing-advisor
- Source: https://github.com/Cloud-Byte-Consulting/plugins
- Type: skill
- Category: other
- tashan score: not scored (catalogued only — too little public evidence)
- Adoption: 9.0
- Upkeep: 95.0
- Freshness: 90.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- License: Apache-2.0
- Official: no

## Install

```sh
cp -r gpu-sharing-advisor ~/.claude/skills/
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-08-22 by tashan (https://tashan.sh) from public evidence. Scorer s5.