OPENAI-COMPATIBLE SERVERLESS INFERENCE ENGINE
Live AI Playground & Telemetry
Execute real-time streaming queries against leading open models. Sub-second TTFT and zero idle GPU cloud costs.
TTFT:9ms
•Throughput:210.4 tok/s
CLUSTER STATUS: PRIVATE BETA (INVITE ONLY)•Bring Your Own Endpoint (vLLM / Ollama) or enter your provisioned API key.
Request Early Access Quota →Temperature0.7
Max Output Tokens512
Welcome to the nexusvllm Live Serverless AI Playground. Select any supported model to execute real-time token streaming, inspect Time-To-First-Token (TTFT), and track usage cost per request.