EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

Test the case when max_new_tokens is very large (#1038)

Browse Source
This commit is contained in:
Lianmin Zheng
2024-08-11 16:41:03 -07:00
committed by GitHub
parent d785412077
commit d84c5e70f7
7 changed files with 100 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/sglang/srt/openai_api/adapter.py
View File

@@ -77,7 +77,7 @@ class FileMetadata:
batch_storage: Dict[str, BatchResponse] = {}
file_id_request: Dict[str, FileMetadata] = {}
file_id_response: Dict[str, FileResponse] = {}
# map file id to file path in SGlang backend
# map file id to file path in SGLang backend
file_id_storage: Dict[str, str] = {}
@@ -335,7 +335,7 @@ async def process_batch(tokenizer_manager, batch_id: str, batch_request: BatchRe
}
except Exception as e:
print("error in SGlang:", e)
print("error in SGLang:", e)
# Update batch status to "failed"
retrieve_batch = batch_storage[batch_id]
retrieve_batch.status = "failed"