336 5 months ago

Qwen3 1.7B (Q8_0) optimized for direct response. Template modified to remove all <think> blocks and force immediate output.

tools thinking
cff3f395ef37 · 120B
{
"repeat_penalty": 1,
"stop": [
"<|im_start|>",
"<|im_end|>"
],
"temperature": 0.6,
"top_k": 20,
"top_p": 0.95
}