BitsExplorer

BitsExplorer finds what people are building with AI, every day. See today's picks →

Preview of Rtx5090 Vllm

Rtx5090 Vllm

Norwegian This item is in Norwegian.

vLLM serving Qwen3-Coder-30B-A3B on RTX 5090 via Docker with OpenAI-compatible API. Auto-restarts on boot, designed for single consumer GPU deployment with 32GB VRAM. Features one model setup, no swap logic, and no keep-alive tuning.

Description excerpted from the original listing, which is linked below.

Open it Source code
Open in English

The English version is the original site passed through Google Translate. Nothing is translated or stored here.

The week in AI-built software

Every Monday: how many projects were built with AI that week, which tools built them, where they came from, and the week's picks.

Read the latest week