unsloth

/

Qwen3.5-4B-GGUF

like 233

Follow

Unsloth AI 21.1k

if we compare qwen3 4b Q6, I have 27 tokens per second, but with qwen3.5 4b Q6 10-11, I use LM studio. What could be the reason for this?
No other 4b models behaved like this Gemma3 4b either