Fix getting the ctx-size for EXL3/EXL2/Transformers models

This commit is contained in:
oobabooga 2025-08-08 17:51:24 -07:00
parent f1147c9926
commit b391ac8eb1

View file

@ -106,9 +106,16 @@ def get_model_metadata(model):
for k in ['max_position_embeddings', 'model_max_length', 'max_seq_len']:
if k in metadata:
model_settings['truncation_length'] = metadata[k]
model_settings['truncation_length_info'] = metadata[k]
model_settings['ctx_size'] = min(metadata[k], 8192)
value = metadata[k]
elif k in metadata.get('text_config', {}):
value = metadata['text_config'][k]
else:
continue
model_settings['truncation_length'] = value
model_settings['truncation_length_info'] = value
model_settings['ctx_size'] = min(value, 8192)
break
if 'rope_theta' in metadata:
model_settings['rope_freq_base'] = metadata['rope_theta']