first commit

app.py

@@ -0,0 +1,34 @@
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+model_id = "LiquidAI/LFM2.5-1.2B-Instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+model = AutoModelForCausalLM.from_pretrained(
+    model_id, device_map="auto", torch_dtype=torch.bfloat16
+)
+
+def chat(message, history):
+    messages = history + [{"role": "user", "content": message}]
+
+    input_ids = tokenizer.apply_chat_template(
+        messages, add_generation_prompt=True,
+        return_tensors="pt", tokenize=True
+    ).to(model.device)
+
+    output = model.generate(
+        input_ids, do_sample=True,
+        temperature=0.1, top_k=50,
+        repetition_penalty=1.05, max_new_tokens=512
+    )
+    return tokenizer.decode(output[0][input_ids.shape[1]:], skip_special_tokens=True)
+
+demo = gr.ChatInterface(
+    fn=chat,
+    type="messages",
+    title="LFM2.5 Chat",
+    description="Chat avec le modèle LiquidAI LFM2.5-1.2B-Instruct",
+)
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,4 @@
+gradio>=4.0.0
+transformers
+torch
+accelerate