[Bugfix][Frontend] Stop beam search on generation config EOS tokens (#62072)
pre-commit / pre-run-check (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s

Signed-off-by: mottopanikeiku <176798723+mottopanikeiku@users.noreply.github.com>
Co-authored-by: mottopanikeiku <176798723+mottopanikeiku@users.noreply.github.com>
This commit is contained in:
alp
2026-10-18 16:49:42 +00:00
committed by GitHub
co-authored by mottopanikeiku
parent fe8f2140ee
commit ff01a6b14f
5 changed files with 110 additions and 16 deletions
+1
View File
@@ -99,6 +99,7 @@ def test_beam_search_abort_returns_partial_outputs_and_continues_other_prompts(
llm = LLM.__new__(LLM)
llm.llm_engine = Mock()
llm.llm_engine.input_processor.generation_config_fields = {}
tokenizer = SimpleNamespace(
eos_token_id=0,
decode=lambda tokens, skip_special_tokens=False: " ".join(