Fix autothink and thinkdeeper for transformers>=5

README.md

@@ -9,9 +9,8 @@
 </p>
 
 <p align="center">
-  <a href="https://github.com/algorithmicsuperintelligence/optillm/stargazers"><img src="https://img.shields.io/github/stars/algorithmicsuperintelligence/optillm?style=social" alt="GitHub stars"></a>
   <a href="https://pypi.org/project/optillm/"><img src="https://img.shields.io/pypi/v/optillm" alt="PyPI version"></a>
-  <a href="https://pypi.org/project/optillm/"><img src="https://img.shields.io/pypi/dm/optillm" alt="PyPI downloads"></a>
+  <a href="https://pepy.tech/projects/optillm"><img src="https://static.pepy.tech/personalized-badge/optillm?period=monthly&units=INTERNATIONAL_SYSTEM&left_color=BLACK&right_color=GREEN&left_text=downloads/month" alt="PyPI Downloads"></a>
   <a href="https://github.com/algorithmicsuperintelligence/optillm/blob/main/LICENSE"><img src="https://img.shields.io/github/license/algorithmicsuperintelligence/optillm" alt="License"></a>
 </p>
 

optillm/__init__.py

@@ -1,5 +1,5 @@
 # Version information
-__version__ = "0.3.21"
+__version__ = "0.3.22"
 
 import os as _os
 

optillm/autothink/processor.py

@@ -243,7 +243,7 @@ def process(self, messages: List[Dict[str, str]]) -> str:
                 thinking_messages,
                 continue_final_message=True,
                 return_tensors="pt"
-            ).to(self.model.device)
+            ).input_ids.to(self.model.device)
 
             # Reset and update token history in steering hooks
             if self.steering_hooks:

optillm/deepconf/processor.py

@@ -108,7 +108,7 @@ def generate_single_trace(self, messages: List[Dict[str, str]],
             messages,
             return_tensors="pt",
             add_generation_prompt=True
-        ).to(self.model.device)
+        ).input_ids.to(self.model.device)
 
         # Initialize generation state
         kv_cache = DynamicCache()

optillm/thinkdeeper.py

@@ -77,8 +77,7 @@ def reasoning_effort(self, messages) -> str:
             messages,
             continue_final_message=True,
             return_tensors="pt"
-        )
-        tokens = tokens.to(self.model.device)
+        ).input_ids.to(self.model.device)
 
         kv = DynamicCache()
         n_thinking_tokens = 0

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "optillm"
-version = "0.3.21"
+version = "0.3.22"
 description = "An optimizing inference proxy for LLMs."
 readme = "README.md"
 license = "Apache-2.0"