NVIDIA
Explore
Models
Blueprints
GPUs
Docs
View All Playbooks
View All Playbooks

onboarding

  • Set Up Local Network Access
  • Open WebUI with Ollama

data science

  • CUDA-X Data Science
  • Optimized JAX
  • Text to Knowledge Graph

tools

  • VS Code
  • DGX Dashboard
  • Comfy UI
  • RAG Application in AI Workbench
  • Set up Tailscale on Your Spark

fine tuning

  • FLUX.1 Dreambooth LoRA Fine-tuning
  • LLaMA Factory
  • Fine-tune with NeMo
  • Fine-tune with Pytorch
  • Unsloth on DGX Spark

use case

  • Vibe Coding in VS Code
  • Build and Deploy a Multi-Agent Chatbot
  • NCCL for Two Sparks
  • Connect Two Sparks
  • Build a Video Search and Summarization (VSS) Agent

inference

  • SGLang Inference Server
  • Multi-modal Inference
  • NIM on Spark
  • NVFP4 Quantization
  • Speculative Decoding
  • TRT LLM for Inference
  • Install and Use vLLM for Inference
Terms of Use
Privacy Policy
Your Privacy Choices
Contact

Copyright © 2025 NVIDIA Corporation

SGLang Inference Server

30 MIN

Install and use SGLang on DGX Spark

View on GitHub
OverviewInstructionsTroubleshooting

Common issues and their resolutions:

SymptomCauseFix
Container fails to start with GPU errorsNVIDIA drivers/toolkit missingInstall nvidia-container-toolkit, restart Docker
Server responds with 404 or connection refusedServer not fully initializedWait 60 seconds, check container logs
Out of memory errors during model loadingInsufficient GPU memoryUse smaller model or increase --tp parameter
Model download failsNetwork connectivity issuesCheck internet connection, retry download
Permission denied accessing /tmpVolume mount issuesUse full path: -v /tmp:/tmp or create dedicated directory

NOTE

DGX Spark uses a Unified Memory Architecture (UMA), which enables dynamic memory sharing between the GPU and CPU. With many applications still updating to take advantage of UMA, you may encounter memory issues even when within the memory capacity of DGX Spark. If that happens, manually flush the buffer cache with:

sudo sh -c 'sync; echo 3 > /proc/sys/vm/drop_caches'

Resources

  • SGLang Documentation
  • DGX Spark Documentation
  • DGX Spark Forum