Upload folder using huggingface_hub

README.md

@@ -1,13 +1,15 @@
 ---
-title: K2 Horizon Demo
-emoji: 🐢
-colorFrom: pink
-colorTo: gray
+title: K2-Horizon-0.9B-GGUF Demo
+emoji: 🧠
+colorFrom: blue
+colorTo: purple
 sdk: gradio
-sdk_version: 6.26.0
-python_version: '3.12'
+sdk_version: 5.50.0
 app_file: app.py
-pinned: false
+short_description: Chat with K2-Horizon 0.9B GGUF model
+python_version: "3.12"
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# K2-Horizon-0.9B-GGUF Demo
+
+A Gradio demo for the [IFM/K2-Horizon-0.9B-GGUF](https://huggingface.co/IFM/K2-Horizon-0.9B-GGUF) model — a compact 0.9B reasoning model supporting English and Chinese.

app.py

@@ -0,0 +1,57 @@
+import spaces
+import torch
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+MODEL_ID = "IFM/K2-Horizon-0.9B-GGUF"
+
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_ID,
+    torch_dtype=torch.float16,
+    device_map="auto",
+    trust_remote_code=True,
+)
+
+
+@spaces.GPU(duration=60)
+def generate(message, history):
+    """Generate a response from the K2-Horizon model."""
+    messages = []
+    for user_msg, bot_msg in history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": bot_msg})
+    messages.append({"role": "user", "content": message})
+
+    inputs = tokenizer.apply_chat_template(
+        messages, tokenize=True, add_generation_prompt=True, return_tensors="pt"
+    ).to(model.device)
+
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs,
+            max_new_tokens=1024,
+            temperature=0.7,
+            top_p=0.9,
+            do_sample=True,
+        )
+
+    response = tokenizer.decode(outputs[0][inputs.shape[1] :], skip_special_tokens=True)
+    return response
+
+
+demo = gr.ChatInterface(
+    fn=generate,
+    title="K2-Horizon-0.9B-GGUF",
+    description="Chat with the K2-Horizon 0.9B reasoning model (GGUF quantized).",
+    examples=[
+        ["Who are you?"],
+        ["What is 84 * 3 / 2?"],
+        ["Tell me an interesting fact about the universe!"],
+        ["Explain quantum computing in simple terms."],
+    ],
+    cache_examples=True,
+)
+
+if __name__ == "__main__":
+    demo.launch(mcp_server=True)

requirements.txt

@@ -0,0 +1,3 @@
+transformers
+accelerate
+sentencepiece