Update processing_qwen2_ts.py to work with the latest vllm patch for ChatTS support.
#16
by
alexanderchemeris
- opened
This change updates the prompt pre-processing to align it with the way vLLM performs prompt updates while maintaining caching capability.
See https://github.com/vllm-project/vllm/pull/16852 for the vllm patch code.
xiezhe24
changed pull request status to
merged