# Coreweave Cost Tuning

> Coreweave Cost Tuning is an Agent Skill for devops & security published in the jeremylongshore/claude-code-plugins-plus-skills repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Optimize CoreWeave GPU cloud costs with right-sizing and scheduling. Use when reducing GPU spend, selecting cost-effective instances, or implementing scale-to-zero for dev workloads. Trigger with phrases like "coreweave cost", "coreweave pricing", "reduce coreweave spend", "coreweave budget".

- Skill name: `coreweave-cost-tuning`
- Category: [DevOps & Security](https://www.openparable.com/skill/devops-security/1)
- Source: https://github.com/jeremylongshore/claude-code-plugins-plus-skills/blob/HEAD/plugins/saas-packs/coreweave-pack/skills/coreweave-cost-tuning/SKILL.md
- Repository: jeremylongshore/claude-code-plugins-plus-skills (2,518 GitHub stars when collected)

## SKILL.md

# CoreWeave Cost Tuning

> **Community-contributed.** Not affiliated with, endorsed by, or sponsored by CoreWeave, Inc. CoreWeave is a registered trademark of CoreWeave, Inc.

## GPU Pricing Reference (approximate)

| GPU | Per GPU/hour | Best For |
|-----|-------------|----------|
| A100 40GB PCIe | ~$1.50 | Development, smaller models |
| A100 80GB PCIe | ~$2.21 | Production inference |
| H100 80GB PCIe | ~$4.76 | High-throughput inference |
| H100 SXM5 (8x) | ~$6.15/GPU | Training, multi-GPU |
| L40 | ~$1.10 | Image generation, light inference |

## Cost Optimization Strategies

### Scale-to-Zero for Dev/Staging

```yaml
autoscaling.knative.dev/minScale: "0"
autoscaling.knative.dev/scaleDownDelay: "5m"
```

### Right-Size GPU Selection

```python
def recommend_gpu(model_size_b: float, inference_only: bool = True) -> str:
    if model_size_b <= 7:
        return "L40" if inference_only else "A100_PCIE_80GB"
    elif model_size_b <= 13:
        return "A100_PCIE_80GB"
    elif model_size_b <= 70:
        return "A100_PCIE_80GB (4x tensor parallel)"
    else:
        return "H100_SXM5 (8x tensor parallel)"
```

### Quantization to Use Smaller GPUs

Use AWQ or GPTQ quantization to fit larger models on smaller GPUs:

```bash
# 70B model at 4-bit fits on single A100-80GB instead of 4x
vllm serve meta-llama/Llama-3.1-70B-Instruct-AWQ --quantization awq
```

## Resources

- [CoreWeave Pricing](https://www.coreweave.com/pricing)
- [CoreWeave GPU Instances](https://docs.coreweave.com/docs/platform/instances/gpu-instances)

## Next Steps

For architecture patterns, see `coreweave-reference-architecture`.

---

Source: https://www.openparable.com/skill/devops-security/coreweave-cost-tuning (OpenParable). Cite the canonical URL above.