From 234aba1c50374dee8d342d4df12e4de9428a544f Mon Sep 17 00:00:00 2001 From: oobabooga <112222186+oobabooga@users.noreply.github.com> Date: Sat, 26 Apr 2025 17:33:47 -0700 Subject: [PATCH] llama.cpp: Simplify the prompt processing progress indicator The progress bar was unreliable --- modules/llama_cpp_server.py | 23 +++-------------------- 1 file changed, 3 insertions(+), 20 deletions(-) diff --git a/modules/llama_cpp_server.py b/modules/llama_cpp_server.py index fb972a32..895f3b1e 100644 --- a/modules/llama_cpp_server.py +++ b/modules/llama_cpp_server.py @@ -11,7 +11,6 @@ from pathlib import Path import llama_cpp_binaries import requests -from tqdm import tqdm from modules import shared from modules.logging_colors import logger @@ -391,31 +390,15 @@ class LlamaServer: def filter_stderr_with_progress(process_stderr): - progress_bar = None progress_pattern = re.compile(r'slot update_slots: id.*progress = (\d+\.\d+)') - try: for line in iter(process_stderr.readline, ''): progress_match = progress_pattern.search(line) - if progress_match: - progress = float(progress_match.group(1)) - - # Create progress bar on first progress message - if progress_bar is None: - progress_bar = tqdm(total=1.0, desc="Prompt Processing", leave=False) - - progress_bar.update(progress - progress_bar.n) - - # Clean up when complete - if progress >= 1.0: - progress_bar.close() - progress_bar = None - - if not line.startswith(('srv ', 'slot ')) and 'log_server_r: request: GET /health' not in line: + sys.stderr.write(line) + sys.stderr.flush() + elif not line.startswith(('srv ', 'slot ')) and 'log_server_r: request: GET /health' not in line: sys.stderr.write(line) sys.stderr.flush() except (ValueError, IOError): - if progress_bar: - progress_bar.close() pass