Update README.md

apply chat template to the reasoning model example

README.md

@@ -1,9 +1,9 @@
----
-license: apache-2.0
-tags:
-- mlx
-base_model: GreenBitAI/DeepSeek-R1-Distill-Qwen-1.5B-layer-mix-bpw-4.0
----
+---
+license: apache-2.0
+tags:
+- mlx
+base_model: GreenBitAI/DeepSeek-R1-Distill-Qwen-1.5B-layer-mix-bpw-4.0
+---
 
 # GreenBitAI/DeepSeek-R1-Distill-Qwen-1.5B-layer-mix-bpw-4.0-mlx
 
@@ -20,5 +20,15 @@ pip install gbx-lm
 from gbx_lm import load, generate
 
 model, tokenizer = load("GreenBitAI/DeepSeek-R1-Distill-Qwen-1.5B-layer-mix-bpw-4.0-mlx")
-response = generate(model, tokenizer, prompt="hello", verbose=True)
+
+prompt = "How can I make an apple cake"
+
+if tokenizer.chat_template is not None:
+    messages = [{"role": "user", "content": prompt}]
+    prompt = tokenizer.apply_chat_template(
+        messages, add_generation_prompt=True
+    )
+    prompt = tokenizer.decode(prompt)
+
+response = generate(model, tokenizer, prompt=prompt, verbose=True, max_tokens=4096)
 ```