versioning: 0.0.6, ux: move buttons, feat: add cloud providers, feat: increese tool call limit

traderai/server.py

@@ -39,6 +39,13 @@ def resource_path(*parts: str) -> Path:
 class ChatRequest(BaseModel):
     message: str
     thread_id: str | None = DEFAULT_THREAD_ID
+    images: list["ChatImageRequest"] = []
+
+
+class ChatImageRequest(BaseModel):
+    name: str = "pasted-image.png"
+    content_type: str = "image/png"
+    image_data: str
 
 
 class ChatThreadRequest(BaseModel):
@@ -114,12 +121,14 @@ def create_app() -> FastAPI:
     plan_runner = ContinualPlanRunner(plan_store, tools, memory)
     tools.plan_runner = plan_runner
     agent = OllamaAgent(
-        settings.ollama_base_url,
-        settings.ollama_model,
+        settings.openai_base_url if settings.model_provider == "openai" else settings.ollama_base_url,
+        settings.openai_model if settings.model_provider == "openai" else settings.ollama_model,
         tools,
         memory=memory,
         user_name=settings.traderai_user_name,
         num_ctx=settings.ollama_num_ctx,
+        provider=settings.model_provider,
+        api_key=settings.openai_api_key,
     )
     plan_runner.bind_agent(agent)
     scheduler.bind_agent(agent)
@@ -171,6 +180,7 @@ def create_app() -> FastAPI:
     async def health() -> dict:
         return {
             "ollama": await agent.health(),
+            "model_provider": settings.model_provider,
             "user": memory.get_profile(),
             "jobs": scheduler.list_jobs(),
             "app_data_dir": settings_payload()["app_data_dir"],
@@ -190,7 +200,19 @@ def create_app() -> FastAPI:
 
     @app.get("/api/ollama/status")
     async def ollama_status() -> dict:
-        return await inspect_ollama()
+        return await inspect_model_provider()
+
+    @app.get("/api/openai/models")
+    async def openai_models() -> dict:
+        status = await inspect_openai()
+        return {
+            "provider": "openai",
+            "configured_model": status.get("configured_model"),
+            "models": status.get("models", []),
+            "message": status.get("message", ""),
+            "detail": status.get("detail", ""),
+            "online": status.get("online", False),
+        }
 
     @app.post("/api/ollama/launch")
     async def launch_ollama() -> dict:
@@ -201,7 +223,7 @@ def create_app() -> FastAPI:
             popen_hidden(command)
         except OSError as exc:
             raise HTTPException(status_code=500, detail=f"Could not launch Ollama: {exc}") from exc
-        status = await inspect_ollama()
+        status = await inspect_model_provider()
         status["message"] = "Ollama launch requested."
         return status
 
@@ -218,7 +240,7 @@ def create_app() -> FastAPI:
             popen_hidden([str(cli), "pull", model])
         except OSError as exc:
             raise HTTPException(status_code=500, detail=f"Could not start model install: {exc}") from exc
-        status = await inspect_ollama()
+        status = await inspect_model_provider()
         status["message"] = f"Started installing model {model}."
         return status
 
@@ -298,14 +320,22 @@ def create_app() -> FastAPI:
     @app.post("/api/chat")
     async def chat(request: ChatRequest) -> dict:
         try:
-            return await agent.chat(request.message, thread_id=request.thread_id)
+            return await agent.chat(
+                request.message,
+                thread_id=request.thread_id,
+                images=[image.model_dump() for image in request.images],
+            )
         except OllamaUnavailable as exc:
             raise HTTPException(status_code=503, detail=str(exc)) from exc
 
     @app.post("/api/chat/stream")
     async def chat_stream(request: ChatRequest) -> StreamingResponse:
         async def events():
-            async for event in agent.chat_events(request.message, thread_id=request.thread_id):
+            async for event in agent.chat_events(
+                request.message,
+                thread_id=request.thread_id,
+                images=[image.model_dump() for image in request.images],
+            ):
                 yield f"data: {json.dumps(event)}\n\n"
 
         return StreamingResponse(events(), media_type="text/event-stream")
@@ -475,6 +505,60 @@ def negotiation_identifier_params(identifier: str) -> dict[str, Any]:
     return {"hash": value}
 
 
+async def inspect_model_provider() -> dict[str, Any]:
+    settings = get_settings()
+    if settings.model_provider == "openai":
+        return await inspect_openai()
+    return await inspect_ollama()
+
+
+async def inspect_openai() -> dict[str, Any]:
+    settings = get_settings()
+    models: list[str] = []
+    online = False
+    detail = ""
+    if not settings.openai_api_key:
+        return {
+            "installed": True,
+            "running": False,
+            "online": False,
+            "provider": "openai",
+            "model_available": False,
+            "configured_model": settings.openai_model,
+            "base_url": settings.openai_base_url,
+            "models": [],
+            "message": "OpenAI is selected, but no API key is configured.",
+            "detail": "",
+        }
+
+    try:
+        async with httpx.AsyncClient(timeout=10) as client:
+            response = await client.get(
+                f"{settings.openai_base_url.rstrip('/')}/models",
+                headers={"Authorization": f"Bearer {settings.openai_api_key}"},
+            )
+        response.raise_for_status()
+        body = response.json()
+        online = True
+        models = sorted(item.get("id") for item in body.get("data", []) if item.get("id"))
+    except (httpx.HTTPError, ValueError) as exc:
+        detail = str(exc)
+
+    model_available = settings.openai_model in models
+    return {
+        "installed": True,
+        "running": online,
+        "online": online,
+        "provider": "openai",
+        "model_available": model_available,
+        "configured_model": settings.openai_model,
+        "base_url": settings.openai_base_url,
+        "models": models,
+        "message": openai_status_message(online, bool(settings.openai_api_key), model_available, settings.openai_model),
+        "detail": detail,
+    }
+
+
 async def inspect_ollama() -> dict[str, Any]:
     settings = get_settings()
     executable = find_ollama_executable()
@@ -500,6 +584,7 @@ async def inspect_ollama() -> dict[str, Any]:
         "installed": installed,
         "running": online,
         "online": online,
+        "provider": "ollama",
         "model_available": model_available,
         "configured_model": settings.ollama_model,
         "base_url": settings.ollama_base_url,
@@ -514,6 +599,16 @@ async def inspect_ollama() -> dict[str, Any]:
     }
 
 
+def openai_status_message(running: bool, configured: bool, model_available: bool, model: str) -> str:
+    if not configured:
+        return "OpenAI API key is not configured."
+    if not running:
+        return "OpenAI is not reachable with the configured key."
+    if not model_available:
+        return f'OpenAI is reachable, but model "{model}" was not returned by the API.'
+    return "OpenAI is ready."
+
+
 def ollama_status_message(installed: bool, running: bool, model_available: bool, model: str) -> str:
     if not installed:
         return "Ollama is not installed."