From e06d8ea22f019e192e29d169b86b0893ca6cd95b Mon Sep 17 00:00:00 2001
From: ShashwatPatil <shashwatpatil974@gmail.com>
Date: Sun, 16 Mar 2025 15:31:51 +0530
Subject: [PATCH] fixed failing node

---
 node-hub/dora-transformer/dora_transformer/main.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/node-hub/dora-transformer/dora_transformer/main.py b/node-hub/dora-transformer/dora_transformer/main.py
index 1fe4ee4d..4c29b859 100644
--- a/node-hub/dora-transformer/dora_transformer/main.py
+++ b/node-hub/dora-transformer/dora_transformer/main.py
@@ -129,8 +129,7 @@ def generate_response(model, tokenizer, text: str, history, max_retries: int = 3
                 else:
                     # Final retry: Reduce token count
                     logging.info("Reducing token count for final attempt")
-                    global MAX_TOKENS
-                    MAX_TOKENS = max(32, MAX_TOKENS // 2)
+                    MAX_TOKENS = 24
                     continue
             else:
                 # For non-CUDA OOM errors, raise immediately