Ollama's a game-changer for democratizing AI - Qwen models run locally and fast, turning scarcity into abundance for coders everywhere. By 2030, expect everyone tinkering with personalized AI like this
What parameter sizes are you talking about? The 30B-A3B is very slow on my device even at low quantizations. The 4B version works, but isn't very good, and the 80B-A3B doesn't even fit!
Ollama's a game-changer for democratizing AI - Qwen models run locally and fast, turning scarcity into abundance for coders everywhere. By 2030, expect everyone tinkering with personalized AI like this
What parameter sizes are you talking about? The 30B-A3B is very slow on my device even at low quantizations. The 4B version works, but isn't very good, and the 80B-A3B doesn't even fit!
https://openrouter.ai/qwen/qwen3-embedding-4b
I am using this through VS Code as of now. works on both home and office.
Wait... This does the job for you? I thought anything sub 10B won't code very well.
Can you tell me what you use it for?