SGLang hosting

SGLang matches or beats vLLM on raw throughput for many workloads and pulls ahead when requests share prefixes (agents, RAG, multi-turn chat) thanks to RadixAttention prefix caching. It also has first-class structured-output (JSON schema) support. DeepSeek and several inference providers run SGLang in production.

Providers with a ready-made SGLang path

ProviderHow
RunPod SGLang template available in template gallery; also common as a Serverless worker base Deploy
Vast.ai Run via official Docker image lmsysorg/sglang on any instance Deploy
Novita AI Deploy the lmsysorg/sglang image on a GPU instance Deploy

Everywhere else: Official image: `docker run --gpus all lmsysorg/sglang:latest python3 -m sglang.launch_server --model-path <hf-repo>`.

Which GPU to rent

Same VRAM math as vLLM. Budget extra RAM if you enable large prefix caches. That is where SGLang earns its keep.

Live floors refreshed 2026-08-23
Model classGPUVRAMFrom $/hrCheapest at
Up to ~17B (FP8) / ~34B (INT4) RTX 4090 24 GB $0.134marketplace Vast.ai Rent
Up to ~23B (FP8) / ~46B (INT4) RTX 5090 32 GB $0.321marketplace Vast.ai Rent
Up to ~34B (FP8) / ~69B (INT4) L40S 48 GB $0.467marketplace Vast.ai Rent
Up to ~58B (FP8) / ~116B (INT4) H100 SXM 80 GB $1.34marketplace Vast.ai Rent
Up to ~102B (FP8) / ~205B (INT4) H200 141 GB $2.00spot Verda
Up to ~130B (FP8) / ~261B (INT4) — MoE giants go multi-GPU B200 180 GB $3.06spot Verda

Rent vs API: when self-hosting SGLang pays off

Self-hosting wins when utilization is high and steady. A rented GPU bills every hour whether you use it or not, while APIs bill per token. Run your numbers in the breakeven calculator, or check any open model's page (e.g. Llama 3.3 70B) for a side-by-side estimate.

Other stacks