[cpburns2009]

Looping is a common problem with the Qwen models. I've had good luck using --repeat-penalty=1.1 with llama.cpp and 27B. vLLM should have a similar option.

[Der_Einzige]

Please switch to using the far superior reptation penalty, DRY. It's built into llamacpp.

[cpburns2009]

What are good DRY setting

[etdznots]

This is the default value!

[cpburns2009]

Llama.cpp defaults to 1.0 (disabled) and so does vLLM. It looks like only ollama defaults to 1.1.