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

[V1][Usage] Refactor speculative decoding configuration and tests (#14434)

Browse Source 
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
This commit is contained in:
shangmingc
2025-03-23 13:28:10 +08:00
committed by GitHub
parent 0661cfef7a
commit 50c9636d87
20 changed files with 1055 additions and 802 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/spec_decode/e2e/conftest.py
View File

@@ -56,7 +56,7 @@ def test_llm_generator(common_llm_kwargs, per_test_common_llm_kwargs,
def maybe_assert_ngram_worker(llm):
# Verify the proposer worker is ngram if ngram is specified.
if (llm.llm_engine.speculative_config is not None
and llm.llm_engine.speculative_config.ngram_prompt_lookup_max > 0):
and llm.llm_engine.speculative_config.method == "ngram"):
from vllm.spec_decode.ngram_worker import NGramWorker
assert isinstance(
llm.llm_engine.model_executor.driver_worker.proposer_worker,