Spaces:

lapa-llm
/

lapa

Running on Zero

robinhad commited on Aug 25

Commit

57c728f

verified ·

1 Parent(s): 21d9138

Update app.py

Files changed (1) hide show

app.py CHANGED Viewed

@@ -9,7 +9,7 @@ import gradio as gr
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
-MODEL_ID = "le-llm/lapa-v0.1-hybrid-reasoning"
@@ -49,7 +49,7 @@ def respond(
         add_generation_prompt=True,
         enable_thinking=True,
     )
-    input_text += "<think>"
     print(input_text)
     inputs = tokenizer(input_text, return_tensors="pt").to(model.device) # .to(device)

 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+MODEL_ID = "le-llm/lapa-v0.1-reasoning-only"
         add_generation_prompt=True,
         enable_thinking=True,
     )
+    input_text += "<think>" # TODO: remove short term fix
     print(input_text)
     inputs = tokenizer(input_text, return_tensors="pt").to(model.device) # .to(device)