Add `skip_special_tokens` sampling params (#1186)

vllm/engine/llm_engine.py

@@ -387,7 +387,7 @@ class LLMEngine:
             child_seqs.append((parent, parent))
 
         for seq, _ in child_seqs:
-            self._decode_sequence(seq)
+            self._decode_sequence(seq, seq_group.sampling_params)
             self._check_stop(seq, seq_group.sampling_params)
 
         # Non-beam search case
@@ -621,7 +621,8 @@ class LLMEngine:
                     f"CPU KV cache usage: {cpu_cache_usage * 100:.1f}%")
         self.last_logging_time = now
 
-    def _decode_sequence(self, seq: Sequence) -> None:
+    def _decode_sequence(self, seq: Sequence,
+                         sampling_params: SamplingParams) -> None:
         """Decodes the new token for a sequence."""
         (new_tokens, new_output_text, prefix_offset,
          read_offset) = detokenize_incrementally(
@@ -630,7 +631,7 @@ class LLMEngine:
              prev_tokens=seq.tokens,
              prefix_offset=seq.prefix_offset,
              read_offset=seq.read_offset,
-             skip_special_tokens=True,
+             skip_special_tokens=sampling_params.skip_special_tokens,
          )
         if seq.tokens is None:
             seq.tokens = new_tokens

vllm/entrypoints/openai/api_server.py

@@ -225,6 +225,7 @@ async def create_chat_completion(request: ChatCompletionRequest,
             top_k=request.top_k,
             ignore_eos=request.ignore_eos,
             use_beam_search=request.use_beam_search,
+            skip_special_tokens=request.skip_special_tokens,
         )
     except ValueError as e:
         return create_error_response(HTTPStatus.BAD_REQUEST, str(e))
@@ -426,6 +427,7 @@ async def create_completion(request: CompletionRequest, raw_request: Request):
             max_tokens=request.max_tokens,
             logprobs=request.logprobs,
             use_beam_search=request.use_beam_search,
+            skip_special_tokens=request.skip_special_tokens,
         )
     except ValueError as e:
         return create_error_response(HTTPStatus.BAD_REQUEST, str(e))

vllm/entrypoints/openai/protocol.py

@@ -71,6 +71,7 @@ class ChatCompletionRequest(BaseModel):
     ignore_eos: Optional[bool] = False
     use_beam_search: Optional[bool] = False
     stop_token_ids: Optional[List[int]] = Field(default_factory=list)
+    skip_special_tokens: Optional[bool] = True
 
 
 class CompletionRequest(BaseModel):
@@ -96,6 +97,7 @@ class CompletionRequest(BaseModel):
     ignore_eos: Optional[bool] = False
     use_beam_search: Optional[bool] = False
     stop_token_ids: Optional[List[int]] = Field(default_factory=list)
+    skip_special_tokens: Optional[bool] = True
 
 
 class LogProbs(BaseModel):

vllm/sampling_params.py

@@ -60,6 +60,8 @@ class SamplingParams:
             tokens after the EOS token is generated.
         max_tokens: Maximum number of tokens to generate per output sequence.
         logprobs: Number of log probabilities to return per output token.
+        skip_special_tokens: Whether to skip special tokens in the output.
+            Defaults to true.
     """
 
     def __init__(
@@ -79,6 +81,7 @@ class SamplingParams:
         ignore_eos: bool = False,
         max_tokens: int = 16,
         logprobs: Optional[int] = None,
+        skip_special_tokens: bool = True,
     ) -> None:
         self.n = n
         self.best_of = best_of if best_of is not None else n
@@ -103,6 +106,7 @@ class SamplingParams:
         self.ignore_eos = ignore_eos
         self.max_tokens = max_tokens
         self.logprobs = logprobs
+        self.skip_special_tokens = skip_special_tokens
 
         self._verify_args()
         if self.use_beam_search:
@@ -196,4 +200,5 @@ class SamplingParams:
                 f"stop={self.stop}, "
                 f"ignore_eos={self.ignore_eos}, "
                 f"max_tokens={self.max_tokens}, "
-                f"logprobs={self.logprobs})")
+                f"logprobs={self.logprobs}, "
+                f"skip_special_tokens={self.skip_special_tokens})")