Verified reference stack
Consultant — Private Chat — RTX 3060
Qwen3-8B is distributed by Ollama at about 5.2 GB; 12 GB VRAM leaves practical headroom for a bounded-context single-user chat.
Open WebUI
Open tool profile →Ollama
Open runtime profile →Qwen3-8B
Open model profile →Fully local chat when only local model providers are enabled
Apache-2.0
Evidence
Continue the decision
Put this reference stack in context
A reference configuration is most useful when you compare it with alternatives and then apply your own hard constraints.
Freshness rule: Material changes set this stack to REVALIDATION_REQUIRED.