Upload 3 files

README.md

@@ -28,4 +28,7 @@ Without `BACKEND_URL`, the Space can still run the same style of workflow locall
 
 - `LLM_BASE_URL`, `LLM_API_KEY`, `LLM_MODEL` for the AMD/vLLM tutor
 - `TRANSLATION_PROVIDER=gemini`, `GEMINI_API_KEY`, `GEMINI_MODEL` for Nepali adaptation and romanized question normalization
+- `TRANSLATION_PROVIDER=openai`, `OPENAI_API_KEY`, `OPENAI_MODEL` if you want to use OpenAI for Nepali adaptation instead
 - `OCR_PROVIDER=gemini`, `OCR_MAX_PAGES=5` for scanned or custom-font PDFs
+
+Use `LLM_MODEL=Qwen/Qwen2.5-7B-Instruct` to match the project default unless your vLLM endpoint exposes a different model name.

app.py

@@ -15,10 +15,12 @@ APP_NAME = os.getenv("APP_NAME", "Pathshala AI")
 BACKEND_URL = os.getenv("BACKEND_URL", "").rstrip("/")
 LLM_BASE_URL = os.getenv("LLM_BASE_URL", "").strip().rstrip("/")
 LLM_API_KEY = os.getenv("LLM_API_KEY", "")
-LLM_MODEL = os.getenv("LLM_MODEL", "Qwen/Qwen2.5-1.5B-Instruct")
+LLM_MODEL = os.getenv("LLM_MODEL", "Qwen/Qwen2.5-7B-Instruct")
 TRANSLATION_PROVIDER = os.getenv("TRANSLATION_PROVIDER", "mock").strip().lower()
 GEMINI_API_KEY = os.getenv("GEMINI_API_KEY", "")
 GEMINI_MODEL = os.getenv("GEMINI_MODEL", "gemini-2.5-flash")
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY", "")
+OPENAI_MODEL = os.getenv("OPENAI_MODEL", "gpt-4o")
 OCR_PROVIDER = os.getenv("OCR_PROVIDER", "off").strip().lower()
 OCR_MAX_PAGES = int(os.getenv("OCR_MAX_PAGES", "5") or "5")
 EMBEDDING_MODEL = os.getenv(
@@ -256,7 +258,19 @@ def adapt_nepali_answer(question, english_answer, sources):
         except (requests.RequestException, KeyError, IndexError, TypeError, ValueError):
             pass
 
-    return nepali_answer(question, " ".join(str(source.get("text", "")) for source in sources))
+    if TRANSLATION_PROVIDER == "openai" and OPENAI_API_KEY:
+        try:
+            translated = translate_with_openai(question, english_answer)
+            translated = remove_source_lines(translated)
+            if is_valid_nepali(translated):
+                return translated
+        except (requests.RequestException, KeyError, IndexError, TypeError, ValueError):
+            pass
+
+    return nepali_answer(
+        question,
+        " ".join(str(source.get("text", "")) for source in sources),
+    )
 
 
 def translate_with_gemini(question, english_answer):
@@ -271,6 +285,46 @@ def translate_with_gemini(question, english_answer):
     return gemini_generate_text(prompt, temperature=0.1, max_output_tokens=450)
 
 
+def translate_with_openai(question, english_answer):
+    response = requests.post(
+        "https://api.openai.com/v1/chat/completions",
+        json={
+            "model": OPENAI_MODEL,
+            "messages": [
+                {
+                    "role": "system",
+                    "content": (
+                        "You adapt English tutoring answers into natural Nepali for "
+                        "primary-school students. Write only Nepali Devanagari. Do not "
+                        "add source labels, markdown, or English sentences."
+                    ),
+                },
+                {
+                    "role": "user",
+                    "content": (
+                        "Translate and simplify this grounded English tutoring answer "
+                        "into natural Nepali for a primary-school student in Nepal. "
+                        "Keep the same meaning. Use Nepali Devanagari only. Do not add "
+                        "new facts. Do not include source citations or headings.\n\n"
+                        f"Student question:\n{question}\n\n"
+                        f"English answer:\n{english_answer}"
+                    ),
+                },
+            ],
+            "temperature": 0.1,
+            "max_tokens": 450,
+        },
+        headers={
+            "Authorization": f"Bearer {OPENAI_API_KEY}",
+            "Content-Type": "application/json",
+        },
+        timeout=45,
+    )
+    response.raise_for_status()
+    data = response.json()
+    return data["choices"][0]["message"]["content"]
+
+
 def normalize_with_gemini(question):
     prompt = (
         "Convert this student question into one clear, simple English question for "
@@ -320,23 +374,53 @@ def fallback_english_answer(sources):
         return "I do not have enough textbook context to answer this question."
 
     topic_text = " ".join(str(source.get("text", "")) for source in sources[:3]).lower()
-    if "soil erosion" in topic_text or "erosion" in topic_text:
+    concept_answer = known_english_concept_answer(topic_text)
+    if concept_answer:
+        return concept_answer
+
+    return "Based on the textbook context, here is the simple explanation: " + truncate(
+        " ".join(context.split()),
+        500,
+    )
+
+
+def known_english_concept_answer(text):
+    if (
+        "living thing" in text
+        or "living things" in text
+        or "organism" in text
+        or "organisms" in text
+    ):
+        return (
+            "Living things are organisms that show the signs of life. They need food "
+            "or energy, breathe or exchange gases, grow, respond to their surroundings, "
+            "and can reproduce. Plants, animals, fungi, and microorganisms are "
+            "examples of living things."
+        )
+
+    if "reflection" in text or "mirror" in text or "image of that object" in text:
+        return (
+            "Reflection of light means light bounces back after hitting a surface. "
+            "A mirror reflects light in an orderly way, so we can see a clear image "
+            "of an object in it. Smooth, flat surfaces make clearer reflections, while "
+            "rough surfaces scatter light and do not show a clear image."
+        )
+
+    if "soil erosion" in text or "erosion" in text:
         return (
             "Soil erosion means the top fertile layer of soil is carried away by "
             "water, wind, or other causes. It makes land less useful for growing "
             "plants, so protecting soil with plants and controlled water flow is important."
         )
-    if "photosynthesis" in topic_text or "chlorophyll" in topic_text:
+
+    if "photosynthesis" in text or "chlorophyll" in text:
         return (
             "Photosynthesis is the process by which green plants make their own food "
             "using sunlight, water, and carbon dioxide. Chlorophyll in leaves helps "
             "plants capture sunlight, and oxygen is released during the process."
         )
 
-    return "Based on the textbook context, here is the simple explanation: " + truncate(
-        " ".join(context.split()),
-        500,
-    )
+    return None
 
 
 def format_sources_for_prompt(sources):
@@ -673,17 +757,122 @@ def normalize_question(question):
             pass
 
     text = cleaned.lower()
-    if "mato" in text and "katan" in text:
+    if has_any(
+        text,
+        [
+            "living thing",
+            "living things",
+            "organism",
+            "organisms",
+            "sajeev",
+            "sajiv",
+            "जीवित",
+            "सजीव",
+        ],
+    ):
+        return "What are living things?"
+
+    if (
+        "soil erosion" in text
+        or "erosion" in text
+        or "माटो कटान" in cleaned
+        or (
+            has_any(text, ["mati", "mato", "matto", "maato"])
+            and has_any(text, ["katan", "katne", "katnu", "bagcha", "bagdai"])
+        )
+    ):
         return "What is soil erosion?"
-    if "prakash" in text and "sansleshan" in text:
+
+    if has_any(text, ["oxygen", "aksijan", "akshijan", "अक्सिजन"]):
+        return "What is oxygen?"
+
+    if (
+        "photosynthesis" in text
+        or "प्रकाश संश्लेषण" in cleaned
+        or (
+            has_any(text, ["prakash", "prakaash"])
+            and has_any(text, ["sansleshan", "samsleshan", "sanshleshan"])
+        )
+    ):
         return "What is photosynthesis?"
-    if "bhinn" in text or "fraction" in text:
+
+    if has_any(text, ["fraction", "bhinn", "vag", "bhaag", "भाग", "भिन्न"]):
         return "What is a fraction?"
+
+    if has_any(text, ["mitochondria", "mitochondrion", "mitokondria"]):
+        return "What is mitochondria?"
+
+    if has_any(text, ["chloroplast", "kloroplast", "chlorophyll"]):
+        return "What is chloroplast?"
+
+    if has_any(text, ["cell", "koshika", "kosika", "कोषिका"]):
+        return "What is a cell?"
+
+    if has_any(text, ["energy", "urja", "oorja", "ऊर्जा"]):
+        return "What is energy?"
+
+    mixed_topic = extract_mixed_language_topic(text)
+    if mixed_topic:
+        return f"What is {mixed_topic}?"
+
     return cleaned
 
 
+def has_any(text, keywords):
+    return any(keyword in text for keyword in keywords)
+
+
+def extract_mixed_language_topic(text):
+    markers = [
+        " vaneko ",
+        " bhaneko ",
+        " vanya ",
+        " bhanya ",
+        " vanne ",
+        " bhanne ",
+    ]
+
+    if not any(marker in f" {text} " for marker in markers):
+        return ""
+
+    topic = f" {text} "
+    removable_phrases = [
+        " vaneko ",
+        " bhaneko ",
+        " vanya ",
+        " bhanya ",
+        " vanne ",
+        " bhanne ",
+        " ke ho ",
+        " k ho ",
+        " kya ho ",
+        " ho ",
+        " ? ",
+    ]
+
+    for phrase in removable_phrases:
+        topic = topic.replace(phrase, " ")
+
+    topic = " ".join(topic.split()).strip(" ?.,")
+    if not topic:
+        return ""
+
+    blocked_words = {"malai", "please", "explain", "bujhau", "bujhaunu", "sir", "mam"}
+    topic_words = [word for word in topic.split() if word not in blocked_words]
+    topic = " ".join(topic_words)
+
+    if not topic or len(topic) > 80:
+        return ""
+
+    return topic
+
+
 def display_topic(question):
     normalized = str(question).lower()
+    if "living thing" in normalized or "organism" in normalized:
+        return "सजीव वस्तु"
+    if "reflection" in normalized:
+        return "प्रकाशको परावर्तन"
     if "photosynthesis" in normalized or "prakash" in normalized:
         return "प्रकाश संश्लेषण"
     if "soil erosion" in normalized or ("mato" in normalized and "katan" in normalized):
@@ -692,22 +881,23 @@ def display_topic(question):
         return "भिन्न"
     if "oxygen" in normalized:
         return "अक्सिजन"
+    if "mitochondria" in normalized or "mitochondrion" in normalized:
+        return "माइटोकन्ड्रिया"
+    if "chloroplast" in normalized:
+        return "क्लोरोप्लास्ट"
+    if "cell" in normalized:
+        return "कोषिका"
+    if "energy" in normalized:
+        return "ऊर्जा"
     return str(question).strip() or "आजको पाठ"
 
 
 def nepali_answer(question, context):
     text = f"{question} {context}".lower()
-    if "soil erosion" in text or "माटो कटान" in context:
-        return (
-            "माटो कटान भनेको पानी, हावा वा अरू कारणले माटोको माथिल्लो मलिलो भाग "
-            "बग्नु वा हट्नु हो। यसले जमिनको उर्वर शक्ति घटाउँछ। रूख, घाँस र बिरुवा "
-            "रोप्दा माटो जोगाउन मद्दत हुन्छ।"
-        )
-    if "photosynthesis" in text or "प्रकाश संश्लेषण" in context:
-        return (
-            "प्रकाश संश्लेषण भनेको हरिया बिरुवाले घामको प्रकाश, पानी र कार्बन "
-            "डाइअक्साइड प्रयोग गरेर खाना बनाउने प्रक्रिया हो। यस क्रममा अक्सिजन पनि निस्कन्छ।"
-        )
+    known_answer = known_nepali_concept_answer(text)
+    if known_answer:
+        return known_answer
+
     if has_devanagari(context):
         return "अपलोड गरिएको पाठ्यपुस्तकको सन्दर्भअनुसार मुख्य कुरा यस्तो छ:\n\n" + truncate(context, 700)
     return (
@@ -716,6 +906,88 @@ def nepali_answer(question, context):
     )
 
 
+def known_nepali_concept_answer(text):
+    if (
+        "living thing" in text
+        or "living things" in text
+        or "organism" in text
+        or "organisms" in text
+        or "sajeev" in text
+        or "sajiv" in text
+        or "सजीव" in text
+        or "जीवित वस्तु" in text
+    ):
+        return (
+            "सजीव वा जीवित वस्तु भनेको जीवनका लक्षण देखाउने वस्तु हो। सजीवले "
+            "खाना वा ऊर्जा लिन्छ, सास फेर्छ, बढ्छ, वातावरणको परिवर्तनमा प्रतिक्रिया "
+            "दिन्छ, र प्रजनन गर्न सक्छ। बिरुवा, जनावर, ढुसी र सूक्ष्म जीवहरू "
+            "सजीवका उदाहरण हुन्।"
+        )
+
+    if "reflection" in text or "mirror" in text or "ऐना" in text or "प्रतिबिम्ब" in text:
+        return (
+            "प्रकाशको परावर्तन भनेको प्रकाश कुनै सतहमा ठोक्किएर फर्कनु हो। ऐनाले "
+            "प्रकाशलाई राम्रोसँग फर्काउँछ, त्यसैले त्यसमा वस्तुको प्रतिबिम्ब देखिन्छ। "
+            "समथर र चिल्लो सतहमा प्रतिबिम्ब प्रस्ट देखिन्छ, तर खस्रो सतहमा प्रकाश धेरै "
+            "दिशामा छरिने भएकाले प्रतिबिम्ब प्रस्ट देखिँदैन।"
+        )
+
+    if "soil erosion" in text or "erosion" in text or "माटो कटान" in text:
+        return (
+            "माटो कटान भनेको हावा, पानी वा अन्य कारणले माटोको माथिल्लो मलिलो भाग "
+            "बिस्तारै बगेर वा उडेर जानु हो। यसले खेतबारीको उर्वर शक्ति घटाउँछ। "
+            "त्यसैले बिरुवा रोप्ने, घाँस जोगाउने र पानीको बहाव नियन्त्रण गर्ने काम "
+            "माटो जोगाउन उपयोगी हुन्छ।"
+        )
+
+    if "oxygen" in text or "अक्सिजन" in text:
+        return (
+            "अक्सिजन एउटा ग्यास हो। जीवित प्राणीले सास फेर्दा अक्सिजन प्रयोग गर्छन्। "
+            "कोषिकाले खाना तोडेर ऊर्जा बनाउन पनि अक्सिजनको मद्दत लिन्छ। "
+            "त्यसैले अक्सिजन जीवनका लागि धेरै महत्त्वपूर्ण हु��्छ।"
+        )
+
+    if "photosynthesis" in text or "chlorophyll" in text or "प्रकाश संश्लेषण" in text:
+        return (
+            "प्रकाश संश्लेषण भनेको हरिया बिरुवाले घामको प्रकाश, पानी र कार्बन डाइअक्साइड "
+            "प्रयोग गरेर आफ्नो खाना बनाउने प्रक्रिया हो। यो काम पातमा हुने हरियो पदार्थ "
+            "क्लोरोफिलको मद्दतले हुन्छ। यस प्रक्रियामा अक्सिजन पनि निस्कन्छ।"
+        )
+
+    if "fraction" in text or "भिन्न" in text:
+        return (
+            "भिन्न भनेको कुनै पूर्ण वस्तुको भाग देखाउने संख्या हो। माथिको संख्या अंश हो, "
+            "जसले कति भाग लिइयो भनेर देखाउँछ। तलको संख्या हर हो, जसले पूर्ण वस्तु कति "
+            "बराबर भागमा बाँडिएको छ भनेर देखाउँछ।"
+        )
+
+    if "mitochondria" in text or "mitochondrion" in text:
+        return (
+            "माइटोकन्ड्रिया कोषिकाभित्र हुने सानो अंगक हो। यसको मुख्य काम खानाबाट ऊर्जा "
+            "बनाउनु हो। त्यसैले यसलाई कोषिकाको ऊर्जा घर पनि भनिन्छ।"
+        )
+
+    if "chloroplast" in text or "plastid" in text:
+        return (
+            "क्लोरोप्लास्ट बिरुवाको कोषिकामा पाइने हरियो अंगक हो। यसमा क्लोरोफिल हुन्छ। "
+            "क्लोरोफिलले घामको प्रकाश लिन मद्दत गर्छ र बिरुवाले खाना बनाउन सक्छ।"
+        )
+
+    if "cell" in text or "कोषिका" in text:
+        return (
+            "कोषिका जीवित वस्तुको सबैभन्दा सानो आधारभूत एकाइ हो। हाम्रो शरीर, बिरुवा "
+            "र धेरै जीवहरू कोषिकाबाट बनेका हुन्छन्। कोषिकाले जीवनका आवश्यक कामहरू गर्छ।"
+        )
+
+    if "energy" in text or "ऊर्जा" in text:
+        return (
+            "ऊर्जा भनेको काम गर्न चाहिने शक्ति हो। जीवित प्राणीले खाना र सास फेर्ने "
+            "प्रक्रियाबाट ऊर्जा पाउँछन्। कोषिकाले यही ऊर्जा प्रयोग गरेर जीवनका काम गर्छ।"
+        )
+
+    return None
+
+
 def nepali_quiz_questions(context):
     short_context = truncate(first_sentence(context), 140)
     return [
@@ -824,6 +1096,8 @@ def startup_status():
     nepali_status = (
         "Gemini Nepali adaptation enabled."
         if TRANSLATION_PROVIDER == "gemini" and GEMINI_API_KEY
+        else "OpenAI Nepali adaptation enabled."
+        if TRANSLATION_PROVIDER == "openai" and OPENAI_API_KEY
         else "Mock Nepali adaptation enabled."
     )
     ocr_status = (