huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 20:34:54 +00:00
Code Issues Packages Projects Releases Wiki Activity

precise comment

Browse Source
This commit is contained in:
fxmarty 2024-06-07 12:43:33 +00:00
parent fb5487d00c
commit b8ac9ba752
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
server/text_generation_server/models/__init__.py
View File

@ -420,7 +420,7 @@ def get_model(
and max_input_tokens > sliding_window
):
raise ValueError(
f"The backend {SYSTEM} does not support sliding window attention. TGI webserver was started max_input_tokens={max_input_tokens} larger than sliding_window={sliding_window}. To use this model with the {SYSTEM} backend, please launch TGI with the argument `--max-batch-prefill-tokens` smaller than {sliding_window}."
f"The backend {SYSTEM} does not support sliding window attention that is used by the model type {model_type}. To use this model nonetheless with the {SYSTEM} backend, please launch TGI with the argument `--max-input-tokens` smaller than sliding_window={sliding_window} (got here max_input_tokens={max_input_tokens})."
)
if model_type == MAMBA: