Refactor chatbot implementation to use local model with transformers. Updated README to reflect new dependencies and usage instructions. Removed Hugging Face Inference API integration.

.gitignore

@@ -0,0 +1,2 @@
+__pycache__/
+.DS_Store

README.md

@@ -9,7 +9,19 @@ app_file: app.py
 pinned: false
 hf_oauth: true
 hf_oauth_scopes:
-- inference-api
+- read-repos
 ---
 
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+An example chatbot using [Gradio](https://gradio.app) and [`transformers`](https://huggingface.co/docs/transformers/index) to run a model locally (downloaded from the Hugging Face Hub).
+
+### Notes
+
+- **Model**: `alibayram/gemma3-tr-v64k-it`
+- **Auth (optional)**: If the model is gated/private, sign in via the app’s Hugging Face login button or set `HF_TOKEN` / `HUGGINGFACEHUB_API_TOKEN` in your environment.
+
+### Local install
+
+```bash
+pip install -U gradio transformers torch
+python app.py
+```

app.py

@@ -1,70 +1,95 @@
+from threading import Thread
+
 import gradio as gr
-from huggingface_hub import InferenceClient
+import torch
+from transformers import (AutoModelForCausalLM, AutoTokenizer,
+                          TextIteratorStreamer)
+
+MODEL_ID = "alibayram/gemma3-tr-v64k-it"
+
+# Model ve tokenizer yükleme
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_ID,
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+)
+
+def build_prompt(sistem_mesaji, gecmis, kullanici_mesaji):
+    mesajlar = []
+
+    if sistem_mesaji:
+        mesajlar.append({
+            "role": "system",
+            "content": sistem_mesaji
+        })
+
+    mesajlar.extend(gecmis)
+
+    mesajlar.append({
+        "role": "user",
+        "content": kullanici_mesaji
+    })
+
+    return tokenizer.apply_chat_template(
+        mesajlar,
+        tokenize=False,
+        add_generation_prompt=True,
+    )
 
 
 def respond(
-    message,
-    history: list[dict[str, str]],
-    system_message,
+    mesaj,
+    gecmis: list[dict[str, str]],
+    sistem_mesaji,
     max_tokens,
     temperature,
     top_p,
-    hf_token: gr.OAuthToken,
 ):
-    """
-    For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-    """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
-
-    messages = [{"role": "system", "content": system_message}]
+    prompt = build_prompt(sistem_mesaji, gecmis, mesaj)
 
-    messages.extend(history)
+    girisler = tokenizer(prompt, return_tensors="pt").to(model.device)
 
-    messages.append({"role": "user", "content": message})
+    streamer = TextIteratorStreamer(
+        tokenizer,
+        skip_prompt=True,
+        skip_special_tokens=True,
+    )
 
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    uretim_parametreleri = dict(
+        **girisler,
+        streamer=streamer,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+        do_sample=True,
+    )
+
+    thread = Thread(target=model.generate, kwargs=uretim_parametreleri)
+    thread.start()
 
-        response += token
-        yield response
+    cevap = ""
+    for token in streamer:
+        cevap += token
+        yield cevap
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
     additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
+        gr.Textbox(
+            value="Sen yardımsever, nazik ve açık şekilde cevap veren bir Türkçe yapay zekâ asistanısın.",
+            label="Sistem Mesajı",
         ),
+        gr.Slider(1, 4096, value=512, step=1, label="Maksimum Yeni Token"),
+        gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Sıcaklık (Temperature)"),
+        gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p"),
     ],
 )
 
 with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
     chatbot.render()
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()

requirements.txt

@@ -0,0 +1,5 @@
+gradio==5.42.0
+transformers>=4.45.0
+torch>=2.2.0
+
+