biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Explicitly set return_dict for apply_chat_template (#33372)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2026-01-30 07:27:04 +00:00
committed by GitHub
parent 726d89720c
commit 9432ed8c7e
11 changed files with 22 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
tests/entrypoints/openai/test_serving_tokens.py
View File

@@ -92,7 +92,8 @@ async def test_same_response_as_chat_completions(client, tokenizer, messages):
messages,
add_generation_prompt=True,
enable_thinking=False, # default with Qwen3
)
return_dict=True, # default with Transformers v5
).input_ids
for ignore_eos in [True, False]:
payload = {
@@ -155,7 +156,8 @@ async def test_stop_string_workflow(client, tokenizer, messages):
messages,
add_generation_prompt=True,
enable_thinking=False, # default with Qwen3
)
return_dict=True, # default with Transformers v5
).input_ids
payload = {
"model": MODEL_NAME,
"token_ids": token_ids,
@@ -251,7 +253,8 @@ async def test_generate_with_lora_adapter(client, tokenizer, messages):
messages,
add_generation_prompt=True,
enable_thinking=False, # default with Qwen3
)
return_dict=True, # default with Transformers v5
).input_ids
payload = {
"model": "Alice",
"token_ids": token_ids,