Update app.py

app.py

@@ -6,56 +6,51 @@ from src.tasks import TaskContext
 from src.utils import pretty_json
 
 
-PROVIDERS = ["openai", "gemini", "deepseek"]
-
-DEFAULTS = {
-    "openai": "gpt-5.2",
-    "gemini": "gemini-3-flash-preview",
-    "deepseek": "deepseek-chat",
-}
+DEFAULT_MODELS = [
+    "openai/gpt-oss-20b",          # OpenAI open-weight :contentReference[oaicite:4]{index=4}
+    "deepseek-ai/DeepSeek-R1",     # DeepSeek :contentReference[oaicite:5]{index=5}
+    "google/gemma-2-2b-it",        # Gemma (open) :contentReference[oaicite:6]{index=6}
+    "google/gemma-3-270m",         # Gemma 3 (multimodal family) :contentReference[oaicite:7]{index=7}
+]
 
 
 def run_pipeline(
-    req_text: str,
-    image_file,  # filepath (str) or None
-
-    analyzer_provider: str, analyzer_model: str, analyzer_temp: float, analyzer_tokens: int,
-    refactor_provider: str, refactor_model: str, refactor_temp: float, refactor_tokens: int,
-    critic_provider: str, critic_model: str, critic_temp: float, critic_tokens: int,
+    req_text, img,
+    analyzer_model, analyzer_backend, analyzer_temp, analyzer_tokens,
+    refactor_model, refactor_backend, refactor_temp, refactor_tokens,
+    critic_model, critic_backend, critic_temp, critic_tokens,
 ):
     if not req_text or not req_text.strip():
         return "Вставь текст требований.", "", "", "{}"
 
-    image_path = image_file if isinstance(image_file, str) else None
-
     cfg = RunConfig(
         analyzer=ModelSpec(
-            provider=analyzer_provider,
             model_id=analyzer_model,
+            backend=analyzer_backend,
             temperature=analyzer_temp,
             max_new_tokens=analyzer_tokens,
-            system_prompt="Отвечай по-русски. Будь точным. Не выдумывай факты. Если не хватает данных — задавай вопросы.",
+            system_prompt="Отвечай по-русски. Будь точным, не выдумывай факты.",
         ),
         refactor=ModelSpec(
-            provider=refactor_provider,
             model_id=refactor_model,
+            backend=refactor_backend,
             temperature=refactor_temp,
             max_new_tokens=refactor_tokens,
-            system_prompt="Отвечай по-русски. Не добавляй требований из головы. Если опираешься на диаграмму — явно укажи, что именно взял.",
+            system_prompt="Отвечай по-русски. Не добавляй требований из головы.",
         ),
         critic=ModelSpec(
-            provider=critic_provider,
             model_id=critic_model,
+            backend=critic_backend,
             temperature=critic_temp,
             max_new_tokens=critic_tokens,
-            system_prompt="Отвечай по-русски. Будь строгим ревьюером. Не прощай двусмысленности и нетестируемые формулировки.",
+            system_prompt="Отвечай по-русски. Будь строгим ревьюером.",
         ),
     )
 
     ctx = TaskContext(name="requirements_analysis_and_refactor", language="ru")
     orch = Orchestrator(cfg, ctx=ctx)
 
-    trace = orch.run(req_text, image_path=image_path)
+    trace = orch.run(req_text, image=img)
     d = trace.to_dict()
 
     return (
@@ -66,44 +61,36 @@ def run_pipeline(
     )
 
 
-with gr.Blocks(title="Multi-agent Requirements Lab (Vision)") as demo:
+with gr.Blocks(title="Multi-agent Requirements Lab") as demo:
     gr.Markdown(
-        "# Multi-agent Requirements Lab (Vision)\n"
-        "Анализ + рефакторинг требований, с поддержкой диаграмм (изображение).\n\n"
-        "## Secrets для Spaces\n"
-        "- `OPENAI_API_KEY`\n"
-        "- `GEMINI_API_KEY`\n"
-        "- `DEEPSEEK_API_KEY`\n"
+        "# Multi-agent Requirements Lab (text + diagram)\n"
+        "LLM-агенты для анализа/рефакторинга требований + извлечение контекста из фото диаграммы.\n\n"
+        "**Для работы через HF Inference API добавь `HF_TOKEN` в Settings → Secrets.**"
     )
 
     with gr.Row():
-        req_text = gr.Textbox(label="Текст требований", lines=14, placeholder="Вставь требования (сырой текст).")
-    with gr.Row():
-        diagram = gr.Image(
-            label="Фото/диаграмма (опционально)",
-            type="filepath",
-        )
+        req_text = gr.Textbox(label="Текст требований", lines=14, placeholder="Вставь требования…")
 
-    gr.Markdown("## Настройки агентов")
+    with gr.Row():
+        img = gr.Image(label="Фото/скрин диаграммы (опционально)", type="pil")
 
-    def agent_controls(title: str, default_provider: str):
+    def agent_controls(title: str, default_model: str):
         with gr.Accordion(title, open=False):
-            provider = gr.Dropdown(PROVIDERS, value=default_provider, label="Provider")
-            model = gr.Textbox(value=DEFAULTS[default_provider], label="Model ID")
+            model = gr.Dropdown(DEFAULT_MODELS, value=default_model, label="Model ID")
+            backend = gr.Radio(["hf_inference_api"], value="hf_inference_api", label="Backend")
             temp = gr.Slider(0.0, 1.5, value=0.2, step=0.05, label="Temperature")
-            tokens = gr.Slider(64, 4000, value=900, step=32, label="Max output tokens")
-        return provider, model, temp, tokens
+            tokens = gr.Slider(64, 2000, value=600, step=32, label="Max new tokens")
+        return model, backend, temp, tokens
 
-    analyzer_provider, analyzer_model, analyzer_temp, analyzer_tokens = agent_controls("Analyzer agent", "gemini")
-    refactor_provider, refactor_model, refactor_temp, refactor_tokens = agent_controls("Refactor agent", "openai")
-    critic_provider, critic_model, critic_temp, critic_tokens = agent_controls("Critic agent", "deepseek")
+    analyzer_model, analyzer_backend, analyzer_temp, analyzer_tokens = agent_controls("Analyzer agent", "openai/gpt-oss-20b")
+    refactor_model, refactor_backend, refactor_temp, refactor_tokens = agent_controls("Refactor agent", "google/gemma-2-2b-it")
+    critic_model, critic_backend, critic_temp, critic_tokens = agent_controls("Critic agent", "deepseek-ai/DeepSeek-R1")
 
     run_btn = gr.Button("Run", variant="primary")
 
-    gr.Markdown("## Результаты")
-    out_an = gr.Markdown()
-    out_rf = gr.Markdown()
-    out_cr = gr.Markdown()
+    out_an = gr.Markdown(label="Analyzer output")
+    out_rf = gr.Markdown(label="Refactor output")
+    out_cr = gr.Markdown(label="Critic output")
 
     with gr.Accordion("Trace (JSON)", open=False):
         trace_json = gr.Code(language="json")
@@ -111,10 +98,10 @@ with gr.Blocks(title="Multi-agent Requirements Lab (Vision)") as demo:
     run_btn.click(
         fn=run_pipeline,
         inputs=[
-            req_text, diagram,
-            analyzer_provider, analyzer_model, analyzer_temp, analyzer_tokens,
-            refactor_provider, refactor_model, refactor_temp, refactor_tokens,
-            critic_provider, critic_model, critic_temp, critic_tokens,
+            req_text, img,
+            analyzer_model, analyzer_backend, analyzer_temp, analyzer_tokens,
+            refactor_model, refactor_backend, refactor_temp, refactor_tokens,
+            critic_model, critic_backend, critic_temp, critic_tokens,
         ],
         outputs=[out_an, out_rf, out_cr, trace_json],
     )