Add local model provider boundary

2026-10-01 13:26:22 +03:00
parent 6b271ad499
commit d459d6c318
3 changed files with 172 additions and 85 deletions
@@ -17,6 +17,7 @@ from fastapi.responses import StreamingResponse
 from pydantic import BaseModel, Field, model_validator

 from app import database
+from app.model_provider import get_model_provider
 from app import workspace

 logger = logging.getLogger("sovereignai.audio")
@@ -58,7 +59,7 @@ class AgentRequest(BaseModel):
    task: str = Field(description="مهمة قصيرة للوكيل المحلي")
    model: str | None = Field(
        default=None,
-        description="اسم نموذج Ollama؛ اتركه فارغًا لاستخدام النموذج الافتراضي",
+        description="اسم النموذج المتاح لدى المزوّد المحلي؛ اتركه فارغًا لاستخدام الافتراضي",
    )


@@ -94,7 +95,7 @@ class ConversationWrite(BaseModel):
 class WebReadRequest(BaseModel):
    url: str = Field(min_length=8, max_length=2048, description="رابط صفحة ويب عامة تريد تحليلها")
    question: str = Field(default="لخّص محتوى الصفحة وأهم نقاطها.", min_length=1, max_length=2000)
-    model: str | None = Field(default=None, description="نموذج Ollama المحلي؛ اتركه فارغًا للنموذج الافتراضي")
+    model: str | None = Field(default=None, description="نموذج المزوّد المحلي؛ اتركه فارغًا للنموذج الافتراضي")


 class _PageText(HTMLParser):
@@ -223,26 +224,20 @@ def validate_conversation_id(value: str) -> str:


 async def get_completion(
-    payload: dict[str, Any], base_url: str, *, timeout_seconds: float = 180.0
+    payload: dict[str, Any], *, timeout_seconds: float = 180.0
 ) -> dict[str, Any]:
-    try:
-        async with httpx.AsyncClient(timeout=timeout_seconds) as client:
-            response = await client.post(f"{base_url}/chat/completions", json=payload)
-            response.raise_for_status()
-            return response.json()
-    except httpx.HTTPStatusError as exc:
-        detail = exc.response.text[:400] or "رفض خادم النموذج الطلب."
-        raise HTTPException(status_code=502, detail=f"خطأ من خادم النموذج المحلي: {detail}") from exc
-    except httpx.RequestError as exc:
-        raise HTTPException(status_code=503, detail="تعذر الاتصال بـ Ollama المحلي على العنوان المضبوط.") from exc
+    provider = get_model_provider()
+    return await provider.complete(payload, timeout_seconds=timeout_seconds)


@app.get("/health")
 def health() -> dict[str, Any]:
+    provider = get_model_provider()
    return {
        "status": "ok",
-        "model": os.getenv("LOCAL_MODEL", "gemma4:e2b"),
-        "backend": os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1"),
+        "provider": provider.name,
+        "model": provider.default_model,
+        "backend": provider.base_url,
        "groq_transcription": "configured" if os.getenv("GROQ_API_KEY") else "not_configured",
        "conversation_database": "sqlite",
        "workspace_agent": "enabled" if workspace.configured_root() else "not_configured",
@@ -251,18 +246,10 @@ def health() -> dict[str, Any]:

@app.get("/v1/models")
 async def list_local_models() -> dict[str, Any]:
-    """List models installed in the configured local Ollama instance."""
-    base_url = os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/")
-    ollama_base = base_url[:-3] if base_url.endswith("/v1") else base_url
-    try:
-        async with httpx.AsyncClient(timeout=10.0) as client:
-            response = await client.get(f"{ollama_base}/api/tags")
-            response.raise_for_status()
-            payload = response.json()
-    except (httpx.HTTPError, ValueError) as exc:
-        raise HTTPException(status_code=503, detail="تعذر جلب قائمة النماذج من Ollama المحلي.") from exc
-    models = [item["name"] for item in payload.get("models", []) if isinstance(item, dict) and item.get("name")]
-    active = os.getenv("LOCAL_MODEL", "gemma4:e2b")
+    """List models available from the configured provider."""
+    provider = get_model_provider()
+    models = await provider.list_models()
+    active = provider.default_model
    if active not in models:
        models.insert(0, active)
    return {"data": [{"id": model, "object": "model"} for model in models]}
@@ -278,7 +265,7 @@ async def ask_workspace(request: WorkspaceAgentRequest) -> dict[str, Any]:
    if not files:
        raise HTTPException(status_code=404, detail="لم أجد نصوصًا مطابقة في ملفات مساحة العمل.")
    context = "\n\n".join(f"--- ملف: {name} ---\n{content}" for name, content in files)
-    model = request.model or os.getenv("LOCAL_MODEL", "gemma4:e2b")
+    model = request.model or get_model_provider().default_model
    payload: dict[str, Any] = {
        "model": model,
        "messages": [
@@ -298,13 +285,7 @@ async def ask_workspace(request: WorkspaceAgentRequest) -> dict[str, Any]:
        ],
        "stream": False,
    }
-    if model.lower().startswith("gemma4"):
-        payload["reasoning_effort"] = "none"
-    completion = await get_completion(
-        payload,
-        os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/"),
-        timeout_seconds=600.0,
-    )
+    completion = await get_completion(payload, timeout_seconds=600.0)
    return {
        "task": request.task,
        "tool": "workspace-search-readonly",
@@ -322,7 +303,7 @@ def get_local_user() -> dict[str, str]:


 def chat_payload(request: ChatRequest, *, stream: bool) -> dict[str, Any]:
-    model = request.model or os.getenv("LOCAL_MODEL", "gemma4:e2b")
+    model = request.model or get_model_provider().default_model
    payload: dict[str, Any] = {
        "model": model,
        "messages": (
@@ -332,7 +313,7 @@ def chat_payload(request: ChatRequest, *, stream: bool) -> dict[str, Any]:
                {
                    "role": "system",
                    "content": (
-                        "أنت مساعد ذكاء اصطناعي محلي يعمل عبر Ollama على جهاز المستخدم. "
+                        "أنت مساعد ذكاء اصطناعي يعمل عبر مزوّد النموذج المحلي المضبوط على جهاز المستخدم. "
                        "أجب بالعربية الواضحة وباختصار مناسب. إذا سُئلت أين أنت، أجب بهذه الصياغة: "
                        "أنا مساعد ذكاء اصطناعي يعمل على جهازك، ولا أملك وجودًا جسديًا أو موقع GPS. "
                        "ولا تدّع معرفة "
@@ -347,54 +328,27 @@ def chat_payload(request: ChatRequest, *, stream: bool) -> dict[str, Any]:
        ),
        "stream": stream,
    }
-    # Disable Gemma 4 thinking so Ollama places the answer in `content` for clients.
-    if model.lower().startswith("gemma4"):
-        payload["reasoning_effort"] = "none"
    return payload


@app.post("/v1/chat/completions")
 async def chat(request: ChatRequest) -> dict[str, Any]:
-    """ترحيل طلب المحادثة إلى النموذج المحلي."""
-    base_url = os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/")
+    """إرسال طلب المحادثة إلى مزوّد النموذج المضبوط."""
    payload = chat_payload(request, stream=False)
-    return await get_completion(payload, base_url)
+    return await get_completion(payload)


@app.post("/v1/chat/stream")
 async def chat_stream(request: ChatRequest) -> StreamingResponse:
-    """Pass Ollama token deltas to clients as newline-delimited JSON."""
-    base_url = os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/")
+    """Pass provider token deltas to clients as newline-delimited JSON."""
+    provider = get_model_provider()
    payload = chat_payload(request, stream=True)

    async def events():
-        try:
-            timeout = httpx.Timeout(connect=15.0, read=None, write=30.0, pool=30.0)
-            async with httpx.AsyncClient(timeout=timeout) as client:
-                async with client.stream(
-                    "POST", f"{base_url}/chat/completions", json=payload
-                ) as response:
-                    if response.status_code >= 400:
-                        detail = (await response.aread()).decode("utf-8", "replace")[:400]
-                        yield json.dumps({"error": f"Ollama {response.status_code}: {detail}"}) + "\n"
-                        return
-                    async for line in response.aiter_lines():
-                        if not line.startswith("data:"):
-                            continue
-                        raw = line[5:].strip()
-                        if raw == "[DONE]":
-                            yield '{"done":true}\n'
-                            return
-                        try:
-                            event = json.loads(raw)
-                            delta = event["choices"][0].get("delta", {}).get("content")
-                        except (ValueError, KeyError, IndexError, TypeError):
-                            continue
-                        if delta:
-                            yield json.dumps({"delta": delta}, ensure_ascii=False) + "\n"
-                    yield '{"done":true}\n'
-        except httpx.RequestError:
-            yield json.dumps({"error": "تعذر الاتصال بـ Ollama المحلي."}, ensure_ascii=False) + "\n"
+        async for event in provider.stream(payload):
+            yield json.dumps(event, ensure_ascii=False) + "\n"
+            if event.get("done") or event.get("error"):
+                return

    return StreamingResponse(
        events(),
@@ -503,8 +457,7 @@ async def run_agent(request: AgentRequest) -> dict[str, Any]:
    except (ValueError, SyntaxError, ZeroDivisionError):
        pass

-    base_url = os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/")
-    model = request.model or os.getenv("LOCAL_MODEL", "gemma4:e2b")
+    model = request.model or get_model_provider().default_model
    payload = {
        "model": model,
        "messages": [
@@ -513,9 +466,7 @@ async def run_agent(request: AgentRequest) -> dict[str, Any]:
        ],
        "stream": False,
    }
-    if model.lower().startswith("gemma4"):
-        payload["reasoning_effort"] = "none"
-    completion = await get_completion(payload, base_url)
+    completion = await get_completion(payload)
    return {
        "task": request.task,
        "tool": "local-llm",
@@ -528,7 +479,7 @@ async def run_agent(request: AgentRequest) -> dict[str, Any]:
 async def read_web_page(request: WebReadRequest) -> dict[str, Any]:
    """Fetch a user-provided public web page and ask the local model about its text."""
    source_url, title, page_text = await _read_public_page(request.url)
-    model = request.model or os.getenv("LOCAL_MODEL", "gemma4:e2b")
+    model = request.model or get_model_provider().default_model
    payload: dict[str, Any] = {
        "model": model,
        "messages": [
@@ -553,11 +504,8 @@ async def read_web_page(request: WebReadRequest) -> dict[str, Any]:
        ],
        "stream": False,
    }
-    if model.lower().startswith("gemma4"):
-        payload["reasoning_effort"] = "none"
    completion = await get_completion(
        payload,
-        os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1").rstrip("/"),
        timeout_seconds=600.0,
    )
    return {
@@ -0,0 +1,139 @@
+"""Provider boundary for local and future model backends."""
+
+from __future__ import annotations
+
+import json
+import os
+from collections.abc import AsyncIterator
+from typing import Any, Protocol
+
+import httpx
+from fastapi import HTTPException
+
+
+class ModelProvider(Protocol):
+    name: str
+    default_model: str
+    base_url: str
+
+    def prepare_payload(self, payload: dict[str, Any]) -> dict[str, Any]: ...
+
+    async def complete(
+        self, payload: dict[str, Any], *, timeout_seconds: float = 180.0
+    ) -> dict[str, Any]: ...
+
+    async def stream(
+        self, payload: dict[str, Any]
+    ) -> AsyncIterator[dict[str, Any]]: ...
+
+    async def list_models(self) -> list[str]: ...
+
+
+class OllamaProvider:
+    name = "ollama"
+
+    def __init__(self, base_url: str, default_model: str) -> None:
+        self.base_url = base_url.rstrip("/")
+        self.default_model = default_model
+
+    def prepare_payload(self, payload: dict[str, Any]) -> dict[str, Any]:
+        prepared = dict(payload)
+        model = str(prepared.get("model", self.default_model))
+        prepared.setdefault("model", model)
+        # Gemma 4 via Ollama otherwise returns its reasoning in a separate field.
+        if model.lower().startswith("gemma4"):
+            prepared.setdefault("reasoning_effort", "none")
+        return prepared
+
+    async def complete(
+        self, payload: dict[str, Any], *, timeout_seconds: float = 180.0
+    ) -> dict[str, Any]:
+        try:
+            async with httpx.AsyncClient(timeout=timeout_seconds) as client:
+                response = await client.post(
+                    f"{self.base_url}/chat/completions",
+                    json=self.prepare_payload(payload),
+                )
+                response.raise_for_status()
+                return response.json()
+        except httpx.HTTPStatusError as exc:
+            detail = exc.response.text[:400] or "رفض خادم النموذج الطلب."
+            raise HTTPException(
+                status_code=502,
+                detail=f"خطأ من خادم النموذج المحلي: {detail}",
+            ) from exc
+        except httpx.RequestError as exc:
+            raise HTTPException(
+                status_code=503,
+                detail="تعذر الاتصال بـ Ollama المحلي على العنوان المضبوط.",
+            ) from exc
+        except ValueError as exc:
+            raise HTTPException(
+                status_code=502, detail="أعاد Ollama استجابة JSON غير صالحة."
+            ) from exc
+
+    async def stream(
+        self, payload: dict[str, Any]
+    ) -> AsyncIterator[dict[str, Any]]:
+        try:
+            timeout = httpx.Timeout(connect=15.0, read=None, write=30.0, pool=30.0)
+            async with httpx.AsyncClient(timeout=timeout) as client:
+                async with client.stream(
+                    "POST",
+                    f"{self.base_url}/chat/completions",
+                    json=self.prepare_payload(payload),
+                ) as response:
+                    if response.status_code >= 400:
+                        detail = (await response.aread()).decode("utf-8", "replace")[:400]
+                        yield {"error": f"Ollama {response.status_code}: {detail}"}
+                        return
+                    async for line in response.aiter_lines():
+                        if not line.startswith("data:"):
+                            continue
+                        raw = line[5:].strip()
+                        if raw == "[DONE]":
+                            yield {"done": True}
+                            return
+                        try:
+                            event = json.loads(raw)
+                            delta = event["choices"][0].get("delta", {}).get("content")
+                        except (ValueError, KeyError, IndexError, TypeError):
+                            continue
+                        if delta:
+                            yield {"delta": delta}
+                    yield {"done": True}
+        except httpx.RequestError:
+            yield {"error": "تعذر الاتصال بـ Ollama المحلي."}
+
+    async def list_models(self) -> list[str]:
+        ollama_base = (
+            self.base_url[:-3] if self.base_url.endswith("/v1") else self.base_url
+        )
+        try:
+            async with httpx.AsyncClient(timeout=10.0) as client:
+                response = await client.get(f"{ollama_base}/api/tags")
+                response.raise_for_status()
+                payload = response.json()
+        except (httpx.HTTPError, ValueError) as exc:
+            raise HTTPException(
+                status_code=503,
+                detail="تعذر جلب قائمة النماذج من Ollama المحلي.",
+            ) from exc
+        return [
+            item["name"]
+            for item in payload.get("models", [])
+            if isinstance(item, dict) and item.get("name")
+        ]
+
+
+def get_model_provider() -> ModelProvider:
+    provider_name = os.getenv("MODEL_PROVIDER", "ollama").strip().lower()
+    if provider_name != "ollama":
+        raise HTTPException(
+            status_code=503,
+            detail=f"مزوّد النموذج '{provider_name}' غير مدعوم حاليًا.",
+        )
+    return OllamaProvider(
+        base_url=os.getenv("LOCAL_LLM_BASE_URL", "http://127.0.0.1:11434/v1"),
+        default_model=os.getenv("LOCAL_MODEL", "gemma4:e2b"),
+    )