fix: correct stream response when enable_thinking is set to false (#5881)

This commit is contained in:
mlmz
2025-05-01 10:44:37 +08:00
committed by GitHub
parent 9f21e75453
commit 256c4c2519
4 changed files with 211 additions and 17 deletions

View File

@@ -69,6 +69,7 @@ DEFAULT_REASONING_MODEL_NAME_FOR_TEST = "deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
DEFAULT_AWQ_MOE_MODEL_NAME_FOR_TEST = (
"hugging-quants/Mixtral-8x7B-Instruct-v0.1-AWQ-INT4"
)
DEFAULT_ENABLE_THINKING_MODEL_NAME_FOR_TEST = "Qwen/Qwen3-30B-A3B"
# Nightly tests
DEFAULT_MODEL_NAME_FOR_NIGHTLY_EVAL_TP1 = "meta-llama/Llama-3.1-8B-Instruct,mistralai/Mistral-7B-Instruct-v0.3,deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct,google/gemma-2-27b-it"