321 1 week ago

Uses the Q8_0 quantized version of the official Qwen/Qwen3.5 model files, without any other modifications.

vision tools thinking 0.8b 2b 4b 9b 27b
9371364b27a5 · 65B
{
"presence_penalty": 1.5,
"temperature": 1,
"top_k": 20,
"top_p": 0.95
}