Remove argument from arg_utils.py

tests/spec_decode/e2e/conftest.py

@@ -282,6 +282,6 @@ def run_greedy_equality_correctness_test(baseline_llm_generator,
         if print_tokens:
             print(f'{i=} {baseline_tokens=}')
             print(f'{i=}     {spec_tokens=}')
-        #print(f'{i=} {baseline_token_ids=}')
-        #print(f'{i=}     {spec_token_ids=}')
+        print(f'{i=} {baseline_token_ids=}')
+        print(f'{i=}     {spec_token_ids=}')
         assert baseline_token_ids == spec_token_ids

tests/spec_decode/e2e/test_multistep_correctness.py

@@ -643,12 +643,12 @@ def test_many_k(baseline_llm_generator, test_llm_generator, batch_size: int,
         # Try a range of common k.
         for k in [1, 2, 3]
     ])
-@pytest.mark.parametrize("batch_size", [1, 8, 64])
+@pytest.mark.parametrize("batch_size", [1, 32])
 @pytest.mark.parametrize(
     "output_len",
     [
         # Use smaller output len for fast test.
-        256,
+        32,
     ])
 @pytest.mark.parametrize("seed", [1])
 def test_typical_acceptance_sampling(baseline_llm_generator,

vllm/config.py

@@ -885,6 +885,7 @@ def maybe_create_spec_config(
             typical_acceptance_sampler_posterior_alpha (Optional[float]):
                 A scaling factor for the entropy-based threshold in the
                 TypicalAcceptanceSampler.
+
         Returns:
             Optional["SpeculativeConfig"]: An instance of SpeculativeConfig if
                 the necessary conditions are met, else None.

vllm/engine/arg_utils.py

@@ -578,7 +578,7 @@ def add_cli_args(parser: FlexibleArgumentParser) -> FlexibleArgumentParser:
             default=EngineArgs.ngram_prompt_lookup_min,
             help='Min size of window for ngram prompt lookup in speculative '
             'decoding.')
-        
+
         parser.add_argument(
             '--spec-decoding-acceptance-method',
             type=str,