The author runs Qwen 3.8 27B at Q6 with a 160k context and vision on an RTX 5090, using the Hermes agent harness and SearxNG in Docker, and reports around 80 tok/s but about 15 minutes for a one-shot website.
Start with your machine.
Capacity is the reported total for this setup. Multi-GPU and unified-memory layouts can behave differently.
A custom 4x RTX 5090 build with 128GB of VRAM, an Intel Xeon w5-3423 and 192GB of ECC DDR5 runs DeepSeek V4 Flash, Qwen 3.8 and Gemma 4 32B locally for a 50-person company that wants its data off grid.
An RTX 5090 runs recurring agents for project maintenance, site contributions and personalized news monitoring.
A legal user has Claude build task plans while a local Qwen 27B executes them, with anonymisation, OCR for difficult documents and a RAG pipeline using nomic embeddings.
A developer with 20 years of experience runs three concurrent coding sessions at about 150 tok/s each on a single RTX 5090 and has been cloud-free for three months.
Three GPUs spread across machines run Qwen 3.8 27B through a custom stateful agent layer to build a personal finance and portfolio tracker without shipping documents to the cloud.
An office replaced roughly $2,000 a month of API consumption with two RTX 5090 machines running Qwen 3.8 27B under vLLM, one for helpdesk and one for ops.
A Python collector pulls read-only configs, firewall rules, and network health data every day and feeds it into a local RAG pipeline, so the advisor always knows the current state of the homelab without ever being able to touch it.