> coreweave-performance-tuning
Optimize CoreWeave GPU inference latency and throughput. Use when reducing inference latency, maximizing GPU utilization, or tuning batch sizes and concurrency. Trigger with phrases like "coreweave performance", "coreweave latency", "coreweave throughput", "optimize coreweave inference".
curl "https://skillshub.wtf/jeremylongshore/claude-code-plugins-plus-skills/coreweave-performance-tuning?format=md"CoreWeave Performance Tuning
GPU Selection by Workload
| Workload | Recommended GPU | Why |
|---|---|---|
| LLM inference (7-13B) | A100 80GB | Good balance of memory and cost |
| LLM inference (70B+) | 8xH100 | NVLink for tensor parallelism |
| Image generation | L40 | Good for diffusion models |
| Training (large models) | 8xH100 SXM5 | Fastest interconnect |
| Batch processing | A100 40GB | Cost-effective |
Inference Optimization
# Continuous batching with vLLM
containers:
- name: vllm
args:
- "--model=meta-llama/Llama-3.1-8B-Instruct"
- "--max-num-batched-tokens=8192"
- "--max-num-seqs=256"
- "--gpu-memory-utilization=0.90"
- "--enable-prefix-caching"
- "--dtype=float16"
Autoscaling Tuning
# HPA based on GPU utilization
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
name: inference-hpa
spec:
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: inference-server
minReplicas: 2
maxReplicas: 10
metrics:
- type: Pods
pods:
metric:
name: DCGM_FI_DEV_GPU_UTIL
target:
type: AverageValue
averageValue: "70"
Performance Benchmarks
| Metric | A100-80GB | H100-80GB |
|---|---|---|
| Llama-8B tokens/sec | ~2,000 | ~4,500 |
| Llama-70B tokens/sec | ~200 (4x) | ~500 (4x) |
| Cold start (vLLM) | 30-60s | 20-40s |
Resources
Next Steps
For cost optimization, see coreweave-cost-tuning.
> related_skills --same-repo
> fathom-cost-tuning
Optimize Fathom API usage and plan selection. Trigger with phrases like "fathom cost", "fathom pricing", "fathom plan".
> fathom-core-workflow-b
Sync Fathom meeting data to CRM and build automated follow-up workflows. Use when integrating Fathom with Salesforce, HubSpot, or custom CRMs, or creating automated post-meeting email summaries. Trigger with phrases like "fathom crm sync", "fathom salesforce", "fathom follow-up", "fathom post-meeting workflow".
> fathom-core-workflow-a
Build a meeting analytics pipeline with Fathom transcripts and summaries. Use when extracting insights from meetings, building CRM sync, or creating automated meeting follow-up workflows. Trigger with phrases like "fathom analytics", "fathom meeting pipeline", "fathom transcript analysis", "fathom action items sync".
> fathom-common-errors
Diagnose and fix Fathom API errors including auth failures and missing data. Use when API calls fail, transcripts are empty, or webhooks are not firing. Trigger with phrases like "fathom error", "fathom not working", "fathom api failure", "fix fathom".