oobabooga/text-generation-webui
1
0
Fork 0
mirror of https://github.com/oobabooga/text-generation-webui.git synced 2026-04-19 13:33:43 +00:00
Code Issues Projects Releases Packages Wiki Activity

llama.cpp: add back the two encode calls (they are harmless now)

Browse source
This commit is contained in:
oobabooga 2025-04-19 17:38:36 -07:00
parent b9da5c7e3a
commit 5ab069786b
2 changed files with 2 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

8
modules/text_generation.py
View file

@ -481,12 +481,8 @@ def generate_reply_custom(question, original_question, seed, state, stopping_str
traceback.print_exc()
finally:
t1 = time.time()
if shared.args.loader == 'llama.cpp':
original_tokens = shared.model.last_input_length
new_tokens = shared.model.last_output_length
else:
original_tokens = len(encode(original_question)[0])
new_tokens = len(encode(original_question + reply)[0]) - original_tokens
original_tokens = len(encode(original_question)[0])
new_tokens = len(encode(original_question + reply)[0]) - original_tokens
print(f'Output generated in {(t1-t0):.2f} seconds ({new_tokens/(t1-t0):.2f} tokens/s, {new_tokens} tokens, context {original_tokens}, seed {seed})')
return