Skip to content

Fix OpenAI Client error with single request via batch api #6170

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mickqian merged 5 commits into from
Jun 5, 2025
Merged

Fix OpenAI Client error with single request via batch api #6170

Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
3345f55
Fix openai client batch with single request
May 10, 2025
b009431
Merge branch 'main' into ravi/fix_batch
ravi03071991 May 12, 2025
842ee23
fix nonetype error
May 12, 2025
e5ce0df
update with suggestions
ravi03071991 May 31, 2025
aa6ac82
Merge branch 'main' into ravi/fix_batch
ravi03071991 May 31, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
50 changes: 40 additions & 10 deletions python/sglang/srt/openai_api/adapter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1213,7 +1213,11 @@ def v1_chat_generate_request(
top_logprobs_num=top_logprobs_nums,
stream=all_requests[0].stream,
return_text_in_logprobs=True,
rid=request_ids,
rid=(
request_ids
if request_ids is None or len(request_ids) > 1
else request_ids[0]
),
modalities=modalities_list,
lora_path=lora_paths,
bootstrap_host=all_requests[0].bootstrap_host,
Expand Down Expand Up @@ -1375,27 +1379,53 @@ def v1_chat_generate_response(
if to_file:
responses = []

for i, choice in enumerate(choices):
if len(choices) == 1:
response = {
"status_code": 200,
"request_id": ret[i]["meta_info"]["id"],
"request_id": ret[0]["meta_info"]["id"],
"body": {
# remain the same but if needed we can change that
"id": ret[i]["meta_info"]["id"],
"id": ret[0]["meta_info"]["id"],
"object": "chat.completion",
"created": created,
"model": request[i].model,
"choices": choice,
"model": request.model,
"choices": choices[0],
"usage": {
"prompt_tokens": ret[i]["meta_info"]["prompt_tokens"],
"completion_tokens": ret[i]["meta_info"]["completion_tokens"],
"total_tokens": ret[i]["meta_info"]["prompt_tokens"]
+ ret[i]["meta_info"]["completion_tokens"],
"prompt_tokens": ret[0]["meta_info"]["prompt_tokens"],
"completion_tokens": ret[0]["meta_info"]["completion_tokens"],
"total_tokens": ret[0]["meta_info"]["prompt_tokens"]
+ ret[0]["meta_info"]["completion_tokens"],
},
"system_fingerprint": None,
},
}

responses.append(response)
else:
for i, choice in enumerate(choices):
response = {
"status_code": 200,
"request_id": ret[i]["meta_info"]["id"],
"body": {
# remain the same but if needed we can change that
"id": ret[i]["meta_info"]["id"],
"object": "chat.completion",
"created": created,
"model": request[i].model,
"choices": choice,
"usage": {
"prompt_tokens": ret[i]["meta_info"]["prompt_tokens"],
"completion_tokens": ret[i]["meta_info"][
"completion_tokens"
],
"total_tokens": ret[i]["meta_info"]["prompt_tokens"]
+ ret[i]["meta_info"]["completion_tokens"],
},
"system_fingerprint": None,
},
}
responses.append(response)

return responses
else:
prompt_tokens = sum(
Expand Down