From 50623cdcd028f9292af4c216d4ddaf0996c1ea89 Mon Sep 17 00:00:00 2001
From: Christian Medina <37550954+cmedinasoriano@users.noreply.github.com>
Date: Fri, 3 Jul 2026 00:14:57 -0400
Subject: [PATCH] fix: use 0-indexing consistently (shard 0-15)

---
 quantize_streaming.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/quantize_streaming.py b/quantize_streaming.py
index 84beac1..eac58e8 100644
--- a/quantize_streaming.py
+++ b/quantize_streaming.py
@@ -47,7 +47,7 @@ def streaming_quantize(model_path: str, output_path: str):
         if (output_path / shard_name).exists():
             return
         
-        print(f"[{idx+1}/{len(shards)}] {Path(shard_file).name} (GPU {gpu_id})")
+        print(f"[{idx}/{len(shards)}] {Path(shard_file).name} (GPU {gpu_id})")
         
         # Load to assigned GPU
         state_dict = load_file(shard_file, device=f"cuda:{gpu_id}")