use 0.5B model + ensure polished UI

app/gradio_app.py

@@ -1,11 +1,7 @@
-"""Gradio chat UI - the easiest deploy target (Hugging Face Spaces).
-
-To deploy on HF Spaces:
-  1. Create a new Space (SDK: Gradio).
-  2. Push this file as `app.py` at the repo root, plus requirements.txt.
-  3. Pick a GPU tier if you want the 7B model; the 1.5B runs on CPU (slowly).
+"""Gradio chat UI — Hugging Face Spaces deploy target.
 
 Locally:  python app/gradio_app.py
+HF Spaces: set app_file: app/gradio_app.py in README front-matter.
 """
 from __future__ import annotations
 
@@ -27,33 +23,53 @@ def _get_assistant():
     return _assistant
 
 
-def respond(message, history, mode, use_repair):
+def respond(message: str, history, mode: str, use_repair: bool) -> str:
     assistant = _get_assistant()
+
     if use_repair:
         from src.agent.repair_loop import make_repair_generator, repair_loop
         gen_fn = make_repair_generator(assistant)
         trace = repair_loop(message, gen_fn, check_program_fn=lambda c: c, max_iters=3)
-        code, note = trace.final_code, f"(repair: {'passed' if trace.success else 'best effort'} in {trace.iterations} iters)"
-        return f"```python\n{code}\n```\n\n_{note}_"
+        note = f"repair: {'passed' if trace.success else 'best effort'} in {trace.iterations} iter(s)"
+        return f"```python\n{trace.final_code}\n```\n\n_{note}_"
 
-    result = assistant.generate(message, mode=mode, return_sources=True)
+    result = assistant.generate(message, mode=mode, max_new_tokens=160, return_sources=True)
     if isinstance(result, tuple):
         code, sources = result
-        src_md = "\n".join(f"- ({s['score']:.2f}) {s['docstring']}" for s in sources)
-        return f"```python\n{code}\n```\n\n**Retrieved examples:**\n{src_md}"
+        src_lines = "\n".join(
+            f"- `{s['score']:.2f}` {s['docstring']}" for s in sources
+        )
+        return f"```python\n{code}\n```\n\n**Retrieved examples**\n{src_lines}"
     return f"```python\n{result}\n```"
 
 
-with gr.Blocks(title="Code Generation Assistant") as demo:
-    gr.Markdown("# Code Generation Assistant\nDescribe what you want; it writes Python, grounded in CodeSearchNet.")
+with gr.Blocks(theme=gr.themes.Soft(), title="Code Generation Assistant") as demo:
+    gr.Markdown(
+        "# Code Generation Assistant\n"
+        "Writes Python from a natural-language request, grounded in retrieved "
+        "CodeSearchNet examples (RAG)."
+    )
     with gr.Row():
-        mode = gr.Radio(["rag", "baseline"], value="rag", label="Mode")
-        use_repair = gr.Checkbox(label="Agentic repair (run + self-fix)", value=False)
+        mode = gr.Radio(
+            ["rag", "baseline"],
+            value="rag",
+            label="Mode",
+            info="RAG conditions the model on retrieved CodeSearchNet examples; baseline is zero-shot.",
+        )
+        use_repair = gr.Checkbox(
+            label="Agentic repair",
+            value=False,
+            info="Generate → run → read error → self-fix loop (up to 3 iterations).",
+        )
     gr.ChatInterface(
         fn=respond,
         additional_inputs=[mode, use_repair],
-        examples=[["Write a function to check if a number is prime."],
-                  ["Parse a CSV file and return a list of dicts."]],
+        examples=[
+            ["Write a function to check if a number is prime."],
+            ["Parse a CSV file and return a list of dictionaries."],
+            ["Reverse a singly linked list."],
+            ["Compute the Levenshtein distance between two strings."],
+        ],
         cache_examples=False,
     )
 

config.yaml

@@ -42,7 +42,7 @@ split:
 
 models:
   embed_model: "sentence-transformers/all-MiniLM-L6-v2"
-  gen_model: "Qwen/Qwen2.5-Coder-1.5B-Instruct"
+  gen_model: "Qwen/Qwen2.5-Coder-0.5B-Instruct"
   top_k: 3
 
 paths: