Upload folder using huggingface_hub

README.md

@@ -1,12 +1,6 @@
 ---
-title: Posttraining Practice
-emoji: ⚡
-colorFrom: yellow
-colorTo: gray
+title: posttraining-practice
+app_file: app.py
 sdk: gradio
 sdk_version: 6.3.0
-app_file: app.py
-pinned: false
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,70 @@
+import gradio as gr
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+
+MODEL_IDS = [
+    "sachiniyer/SmolLM2-DPO-Schwinn-SmolLM2-Base",
+]
+
+# Load all models
+models = {}
+for model_id in MODEL_IDS:
+    print(f"Loading model: {model_id}")
+    tokenizer = AutoTokenizer.from_pretrained(model_id)
+    model = AutoModelForCausalLM.from_pretrained(
+        model_id,
+        torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+        device_map="auto" if torch.cuda.is_available() else None,
+    )
+    models[model_id] = {"model": model, "tokenizer": tokenizer}
+    print(f"Loaded: {model_id}")
+
+
+def make_respond_fn(model_id: str):
+    def respond(message: str, history: list[tuple[str, str]]) -> str:
+        tokenizer = models[model_id]["tokenizer"]
+        model = models[model_id]["model"]
+
+        # Build conversation from history
+        conversation = ""
+        for user_msg, assistant_msg in history:
+            conversation += f"User: {user_msg}\nAssistant: {assistant_msg}\n"
+        conversation += f"User: {message}\nAssistant:"
+
+        inputs = tokenizer(conversation, return_tensors="pt")
+        if torch.cuda.is_available():
+            inputs = inputs.to("cuda")
+
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=256,
+            do_sample=True,
+            temperature=0.7,
+            top_p=0.9,
+            pad_token_id=tokenizer.eos_token_id,
+        )
+
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        # Extract only the new assistant response
+        response = response.split("Assistant:")[-1].strip()
+        return response
+
+    return respond
+
+
+# Create tabbed interface with one chat per model
+with gr.Blocks(title="posttraining-practice") as demo:
+    gr.Markdown("# posttraining-practice")
+    gr.Markdown("Chat with different fine-tuned models")
+
+    with gr.Tabs():
+        for model_id in MODEL_IDS:
+            short_name = model_id.split("/")[-1]
+            with gr.Tab(short_name):
+                gr.ChatInterface(
+                    fn=make_respond_fn(model_id),
+                    description=f"Chatting with: {model_id}",
+                )
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,434 @@
+# This file was autogenerated by uv via the following command:
+#    uv export --group site --no-hashes --no-dev
+accelerate==1.12.0
+    # via
+    #   peft
+    #   posttraining-study
+    #   trl
+aiofiles==24.1.0
+    # via gradio
+aiohappyeyeballs==2.6.1
+    # via aiohttp
+aiohttp==3.13.2
+    # via
+    #   fsspec
+    #   modal
+aiosignal==1.4.0
+    # via aiohttp
+annotated-doc==0.0.4
+    # via fastapi
+annotated-types==0.7.0
+    # via pydantic
+anyio==4.12.0
+    # via
+    #   gradio
+    #   httpx
+    #   openai
+    #   starlette
+    #   watchfiles
+attrs==25.4.0
+    # via aiohttp
+audioop-lts==0.2.2 ; python_full_version >= '3.13'
+    # via gradio
+beautifulsoup4==4.14.3
+    # via posttraining-study
+bitsandbytes==0.49.0
+    # via posttraining-study
+brotli==1.2.0
+    # via gradio
+cbor2==5.7.1
+    # via modal
+certifi==2025.11.12
+    # via
+    #   httpcore
+    #   httpx
+    #   modal
+    #   requests
+    #   sentry-sdk
+charset-normalizer==3.4.4
+    # via requests
+click==8.3.1
+    # via
+    #   modal
+    #   typer
+    #   uvicorn
+    #   wandb
+colorama==0.4.6 ; sys_platform == 'win32'
+    # via
+    #   click
+    #   tqdm
+datasets==4.4.1
+    # via
+    #   distilabel
+    #   posttraining-study
+    #   trl
+dill==0.4.0
+    # via
+    #   datasets
+    #   multiprocess
+distilabel==1.5.3
+    # via posttraining-study
+distro==1.9.0
+    # via openai
+fastapi==0.128.0
+    # via gradio
+ffmpy==1.0.0
+    # via gradio
+filelock==3.20.0
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   torch
+    #   transformers
+frozenlist==1.8.0
+    # via
+    #   aiohttp
+    #   aiosignal
+fsspec==2025.10.0
+    # via
+    #   datasets
+    #   gradio-client
+    #   huggingface-hub
+    #   torch
+    #   universal-pathlib
+gitdb==4.0.12
+    # via gitpython
+gitpython==3.1.46
+    # via wandb
+gradio==6.3.0
+gradio-client==2.0.3
+    # via gradio
+groovy==0.1.2
+    # via gradio
+grpclib==0.4.8
+    # via modal
+h11==0.16.0
+    # via
+    #   httpcore
+    #   uvicorn
+h2==4.3.0
+    # via grpclib
+hf-xet==1.2.0 ; platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'
+    # via huggingface-hub
+hpack==4.1.0
+    # via h2
+httpcore==1.0.9
+    # via httpx
+httpx==0.28.1
+    # via
+    #   datasets
+    #   distilabel
+    #   gradio
+    #   gradio-client
+    #   openai
+    #   safehttpx
+huggingface-hub==0.36.0
+    # via
+    #   accelerate
+    #   datasets
+    #   gradio
+    #   gradio-client
+    #   peft
+    #   tokenizers
+    #   transformers
+hyperframe==6.1.0
+    # via h2
+idna==3.11
+    # via
+    #   anyio
+    #   httpx
+    #   requests
+    #   yarl
+jinja2==3.1.6
+    # via
+    #   distilabel
+    #   gradio
+    #   torch
+jiter==0.12.0
+    # via openai
+markdown-it-py==4.0.0
+    # via rich
+markupsafe==3.0.3
+    # via
+    #   gradio
+    #   jinja2
+mdurl==0.1.2
+    # via markdown-it-py
+modal==1.2.4
+    # via posttraining-study
+mpmath==1.3.0
+    # via sympy
+multidict==6.7.0
+    # via
+    #   aiohttp
+    #   grpclib
+    #   yarl
+multiprocess==0.70.18
+    # via
+    #   datasets
+    #   distilabel
+nest-asyncio==1.6.0
+    # via distilabel
+networkx==3.6
+    # via
+    #   distilabel
+    #   torch
+numpy==2.3.5
+    # via
+    #   accelerate
+    #   bitsandbytes
+    #   datasets
+    #   gradio
+    #   pandas
+    #   peft
+    #   scipy
+    #   transformers
+nvidia-cublas-cu12==12.8.4.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via
+    #   nvidia-cudnn-cu12
+    #   nvidia-cusolver-cu12
+    #   torch
+nvidia-cuda-cupti-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cuda-nvrtc-cu12==12.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cuda-runtime-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cudnn-cu12==9.10.2.21 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cufft-cu12==11.3.3.83 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cufile-cu12==1.13.1.3 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-curand-cu12==10.3.9.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cusolver-cu12==11.7.3.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-cusparse-cu12==12.5.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via
+    #   nvidia-cusolver-cu12
+    #   torch
+nvidia-cusparselt-cu12==0.7.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-nccl-cu12==2.27.5 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-nvjitlink-cu12==12.8.93 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via
+    #   nvidia-cufft-cu12
+    #   nvidia-cusolver-cu12
+    #   nvidia-cusparse-cu12
+    #   torch
+nvidia-nvshmem-cu12==3.3.20 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+nvidia-nvtx-cu12==12.8.90 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+openai==2.15.0
+    # via posttraining-study
+orjson==3.11.5
+    # via
+    #   distilabel
+    #   gradio
+packaging==25.0
+    # via
+    #   accelerate
+    #   bitsandbytes
+    #   datasets
+    #   gradio
+    #   gradio-client
+    #   huggingface-hub
+    #   peft
+    #   transformers
+    #   wandb
+pandas==2.3.3
+    # via
+    #   datasets
+    #   gradio
+pathlib-abc==0.5.2
+    # via universal-pathlib
+peft==0.18.0
+    # via posttraining-study
+pillow==12.1.0
+    # via
+    #   gradio
+    #   posttraining-study
+platformdirs==4.5.1
+    # via wandb
+portalocker==3.2.0
+    # via distilabel
+propcache==0.4.1
+    # via
+    #   aiohttp
+    #   yarl
+protobuf==6.33.2
+    # via
+    #   modal
+    #   wandb
+psutil==7.1.3
+    # via
+    #   accelerate
+    #   peft
+pyarrow==22.0.0
+    # via datasets
+pydantic==2.12.5
+    # via
+    #   distilabel
+    #   fastapi
+    #   gradio
+    #   openai
+    #   wandb
+pydantic-core==2.41.5
+    # via pydantic
+pydub==0.25.1
+    # via gradio
+pygments==2.19.2
+    # via rich
+python-dateutil==2.9.0.post0
+    # via pandas
+python-dotenv==1.2.1
+    # via posttraining-study
+python-multipart==0.0.21
+    # via gradio
+pytz==2025.2
+    # via pandas
+pywin32==311 ; sys_platform == 'win32'
+    # via portalocker
+pyyaml==6.0.3
+    # via
+    #   accelerate
+    #   datasets
+    #   gradio
+    #   huggingface-hub
+    #   peft
+    #   posttraining-study
+    #   transformers
+    #   wandb
+regex==2025.11.3
+    # via transformers
+requests==2.32.5
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   transformers
+    #   wandb
+rich==14.2.0
+    # via
+    #   distilabel
+    #   modal
+    #   typer
+safehttpx==0.1.7
+    # via gradio
+safetensors==0.7.0
+    # via
+    #   accelerate
+    #   peft
+    #   transformers
+scipy==1.16.3
+    # via distilabel
+semantic-version==2.10.0
+    # via gradio
+sentry-sdk==2.49.0
+    # via wandb
+setuptools==80.9.0
+    # via
+    #   distilabel
+    #   torch
+shellingham==1.5.4
+    # via typer
+six==1.17.0
+    # via python-dateutil
+smmap==5.0.2
+    # via gitdb
+sniffio==1.3.1
+    # via openai
+soupsieve==2.8.1
+    # via beautifulsoup4
+starlette==0.50.0
+    # via
+    #   fastapi
+    #   gradio
+sympy==1.14.0
+    # via torch
+synchronicity==0.10.5
+    # via modal
+tblib==3.2.2
+    # via distilabel
+tenacity==9.1.2
+    # via posttraining-study
+tokenizers==0.22.1
+    # via transformers
+toml==0.10.2
+    # via modal
+tomlkit==0.13.3
+    # via gradio
+torch==2.9.1
+    # via
+    #   accelerate
+    #   bitsandbytes
+    #   distilabel
+    #   peft
+    #   posttraining-study
+tqdm==4.67.1
+    # via
+    #   datasets
+    #   huggingface-hub
+    #   openai
+    #   peft
+    #   transformers
+transformers==4.57.3
+    # via
+    #   distilabel
+    #   peft
+    #   posttraining-study
+    #   trl
+triton==3.5.1 ; platform_machine == 'x86_64' and sys_platform == 'linux'
+    # via torch
+trl==0.25.1
+    # via posttraining-study
+typer==0.20.0
+    # via
+    #   distilabel
+    #   gradio
+    #   modal
+types-certifi==2021.10.8.3
+    # via modal
+types-toml==0.10.8.20240310
+    # via modal
+typing-extensions==4.15.0
+    # via
+    #   aiosignal
+    #   anyio
+    #   beautifulsoup4
+    #   fastapi
+    #   gradio
+    #   gradio-client
+    #   huggingface-hub
+    #   modal
+    #   openai
+    #   pydantic
+    #   pydantic-core
+    #   starlette
+    #   synchronicity
+    #   torch
+    #   typer
+    #   typing-inspection
+    #   wandb
+typing-inspection==0.4.2
+    # via pydantic
+tzdata==2025.2
+    # via pandas
+universal-pathlib==0.3.7
+    # via distilabel
+urllib3==2.6.0
+    # via
+    #   requests
+    #   sentry-sdk
+uvicorn==0.40.0
+    # via gradio
+wandb==0.24.0
+    # via posttraining-study
+watchfiles==1.1.1
+    # via modal
+xxhash==3.6.0
+    # via datasets
+yarl==1.22.0
+    # via aiohttp