sgl-project/sglang · error · ValueError
return_token_ids is not supported with streaming on…
Error message
return_token_ids is not supported with streaming on /v1/chat/completions. Please set stream=false when using return_token_ids=true.
What it means
OpenAI-compatible serving error: returning generated token ids is unsupported with streaming on /v1/chat/completions; set stream=false to use return_token_ids.
Solutions
- Set stream=false to receive generated token IDs
- Or remove return_token_ids when streaming
Example fix
// before
{"stream": true, "return_token_ids": true}
// after
{"stream": false, "return_token_ids": true} Defensive patterns
Strategy: validation
Validate before calling
if body.get("stream"):
body["return_token_ids"] = False
# or assert not (stream and return_token_ids) Type guard
def stream_compat_ok(b): return not (b.get('stream') and b.get('return_token_ids')) Prevention
- Keep a lint rule: stream=true excludes all return_* debug flags
When it happens
Trigger: POST /v1/chat/completions with {"stream": true, "return_token_ids": true}.
Common situations: Debugging tokenization during streaming demos; reusing a non-streaming request payload with stream flipped to true.
Related errors
- return_prompt_token_ids is not supported with streaming…
- return_meta_info is not supported with streaming. Please…
- Assistant tool call function.arguments must be a JSON…
- Assistant tool call function.arguments must be valid JSON.
- Exactly one of 'prompt' or 'messages' must be provided.
AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28).
Data as JSON: /api/errors/d4ecfc2aa2fe4a28.
Report an issue: GitHub.
Appendix: source
Thrown at python/sglang/srt/entrypoints/openai/serving_chat.py:991
if request.chat_template_kwargs
else None
)
if self.is_gpt_oss and reasoning_effort == "none":
raise ValueError(
f"Harmony does not support reasoning effort {reasoning_effort}"
)
if reasoning_effort is not None:
request.reasoning_effort = reasoning_effort
if request.stream:
if request.return_prompt_token_ids:
raise ValueError(
"return_prompt_token_ids is not supported with streaming. "
"Please set stream=false when using return_prompt_token_ids=true."
)
if request.return_token_ids:
raise ValueError(
"return_token_ids is not supported with streaming on "
"/v1/chat/completions. Please set stream=false when using "
"return_token_ids=true."
)
if request.return_meta_info:
raise ValueError(
"return_meta_info is not supported with streaming. "
"Please set stream=false when using return_meta_info=true."
)
is_multimodal = self.tokenizer_manager.model_config.is_multimodal
# Process messages and apply chat template
processed_messages = self._process_messages(request, is_multimodal)
# Build sampling parameters
sampling_params = request.to_sampling_params(
stop=processed_messages.stop,
model_generation_config=self.default_sampling_params,View on GitHub (pinned to 0132848349)