Skip to main content

Self-Serve Deployments

Self-serve deployments provide reserved inference capacity on Crusoe's optimized inference engine and managed infrastructure. They give you predictable performance, dedicated throughput, and scalability that grows with your workload.

When to use self-serve deployments

Self-serve deployments are best when you need:

  • Predictability — Consistent, predictable inference performance over time
  • Sustained traffic — High-volume processing that benefits from reduced cost-per-token as you scale
  • Fine-tuned models — Deployment of your own Low-Rank Adaptation (LoRA) adapters trained through Serverless Fine-Tuning offering
  • Control over scaling — Reduced risk of rate limiting and direct control over the number of replicas supporting your workload
  • Pay-per-use flexibility — GPU-hour billing that lets you align cost with actual usage

Self-serve vs. serverless

If your workload is sporadic or low-volume, serverless might be a better fit. Use the following table to compare the two options across the dimensions that most affect cost and performance.

DimensionSelf-serveServerless
CapacityDedicated deployment with reserved GPUsShared multi-tenant pool
Rate limitsNone; throughput bounded by replica countShared pool limits
BillingGPU-hour (time-based)Per-token
Model supportBase models and LoRA adapters from Serverless Fine-TuningBase models only; no custom or fine-tuned models
Cost efficiencyBest at sustained high utilizationBest for sporadic or low-volume use

Choose an optimization profile

Every self-serve deployment is configured with one optimization profile per model. The profile determines how the inference engine is tuned for your workload.

ProfileOptimizationBest for
ResponsivenessLow latency, optimized for quick responsesInteractive applications, real-time inference, latency-sensitive workloads
ThroughputCost efficiency at scale, optimized for token volumeBatch processing, high-volume workflows, cost-per-token minimization
BalancedHybrid blend of throughput and responsiveness, optimized to support moderate token volume and latencyGeneral purpose production traffic

Supported models

Self-serve deployments support the following base models. You can also deploy these models with LoRA adapters trained through Serverless Fine-Tuning.

LabModelInput modalitiesOutput modalities
DeepSeekDeepSeek V4 FlashTextText
GoogleGemma 4 31B ITImage, TextText
MetaLlama 3.1 8B InstructTextText
MetaLlama 3.3 70B InstructTextText
Moonshot AIKimi K2.6Text, ImageText
OpenAIGPT-OSS 20BTextText
OpenAIGPT-OSS 120BTextText
QwenQwen3 8BTextText
QwenQwen3 235B A22B InstructTextText
QwenQwen3.5 2BImage, TextText
QwenQwen3.5 9BImage, TextText
QwenQwen3.6 27BImage, TextText
QwenQwen3.6 35BImage, TextText
Z AIGLM 5.1TextText

Next steps