feat: 模型推理端到端闭环 — 真实流式推理 + 释放/删除 + GPU 状态同步

后端 (platform.py + platform_store.py):
- 新增 _build_messages_payload() 转换前端格式为 OpenAI messages
- 新增 _stream_chat_proxy() SSE 流式代理到算力节点
- 新增 _unload_from_compute_node() 真正释放算力节点 GPU 显存
- 重写 model_compare_load: 从假 PID/端口改为真正调用算力节点加载模型
- 修复 model_compare_unload: 调用 _unload_from_compute_node 释放 GPU
- 修复 model_compare_delete: 先释放 GPU 再删除记录
- 修复 model_compare_stream_chat: 从 mock 改为 StreamingResponse 代理
- 修复 model_chat_local/stream: 消息格式转换 + 路径修正
- PlatformStore 新增 _inference_nodes 追踪，gpus() 同步推理占用状态
- preload/unload 端点标记/清除推理节点占用

算力节点 (compute):
- inference.py: 适配新版 LLaMA-Factory API (get_infer_args 4 返回值、ChatModel args dict、stream_chat 新签名)
- inference.py: unload() 增加 gc.collect + torch.cuda.empty_cache + synchronize 彻底释放显存
- main.py: inference/load 移除 HTTPException(500)，错误以 200 正常返回

前端:
- InferenceChatView: 真实模式下走 SSE 流式推理，mock 模式保留兼容
- InferenceCreateView: 调用 preloadLocalModel + createCompare 真实创建推理任务，失败回退 mock
- InferenceListView: 「停止」改为「释放」，删除前先释放算力节点，改进错误提示
- compare.ts: 新增 streamChatReal() fetch SSE，preload 超时提升至 5 分钟
- useStreamChat.ts: send() 支持 useMock 参数，真实模式调用 streamChatReal
- GPU 选择过滤: 仅显示在线算力节点上的空闲 GPU

Co-Authored-By: Claude <noreply@anthropic.com>

backend/app/api/v1/endpoints/platform.py

@@ -31,6 +31,62 @@ def _select_first_online_node(store: Any) -> dict[str, Any] | None:
     return None
 
 
+def _build_messages_payload(payload: dict[str, Any]) -> dict[str, Any]:
+    """Convert frontend inference payload to compute API messages format.
+
+    Accepts both:
+      - OpenAI-style: {messages: [{role, content}, ...], temperature, ...}
+      - Frontend-style: {user_question, system_prompt, temperature, ...}
+    """
+    if payload.get("messages"):
+        messages = payload["messages"]
+        # messages already in OpenAI format; pass through with optional system prompt
+        if payload.get("system_prompt") and not any(m.get("role") == "system" for m in messages):
+            messages = [{"role": "system", "content": payload["system_prompt"]}] + list(messages)
+    else:
+        messages = []
+        if payload.get("system_prompt"):
+            messages.append({"role": "system", "content": payload["system_prompt"]})
+        question = payload.get("user_question") or payload.get("question") or ""
+        if question:
+            messages.append({"role": "user", "content": question})
+    return {
+        "messages": messages,
+        "temperature": float(payload.get("temperature", 0.7)),
+        "top_p": float(payload.get("top_p", 0.95)),
+        "max_new_tokens": int(payload.get("max_tokens", 2048)),
+        "do_sample": bool(payload.get("do_sample", True)),
+    }
+
+
+async def _stream_chat_proxy(payload: dict[str, Any]) -> StreamingResponse:
+    """Common SSE streaming proxy: convert payload → forward to compute node → stream back."""
+    store = get_platform_store()
+    node = _select_first_online_node(store)
+    if not node:
+        return StreamingResponse(
+            iter(['data: {"error": "no online compute node available for inference"}\n\n']),
+            media_type="text/event-stream",
+        )
+    client = ComputeNodeClient(node["api_base_url"])
+    compute_payload = _build_messages_payload(payload)
+
+    async def stream_proxy():
+        async with httpx.AsyncClient(timeout=300) as http:
+            url = f"{node['api_base_url'].rstrip('/')}{client.route_prefix}/inference/chat/stream"
+            try:
+                async with http.stream("POST", url, json=compute_payload, headers=client.headers()) as resp:
+                    if resp.status_code >= 400:
+                        yield f'data: {{"error": "compute node returned {resp.status_code}"}}\n\n'.encode()
+                        return
+                    async for chunk in resp.aiter_bytes():
+                        yield chunk
+            except Exception as exc:
+                yield f'data: {{"error": "stream proxy failed: {exc}"}}\n\n'.encode()
+
+    return StreamingResponse(stream_proxy(), media_type="text/event-stream")
+
+
 def fail(status_code: int, message: str) -> HTTPException:
     return HTTPException(status_code=status_code, detail={"code": status_code, "message": message, "data": None})
 
@@ -1022,8 +1078,27 @@ async def model_compare_detail(task_id: str) -> dict[str, Any]:
         raise fail(404, "compare task not found")
 
 
+async def _unload_from_compute_node() -> dict[str, Any]:
+    """Best-effort unload the inference model from the first online compute node."""
+    store = get_platform_store()
+    # Clear all inference tracking — only one model can be loaded at a time
+    for node in store.compute_nodes():
+        store.mark_inference_unloaded(node["id"])
+    node = _select_first_online_node(store)
+    if not node:
+        return {"unloaded": False, "error": "no online compute node"}
+    try:
+        client = ComputeNodeClient(node["api_base_url"])
+        result = await client._request("POST", "/inference/unload", json_data={})
+        return result
+    except Exception as exc:
+        return {"unloaded": False, "error": str(exc)}
+
+
 @router.delete("/model-compare/{task_id}")
 async def model_compare_delete(task_id: str) -> dict[str, Any]:
+    # 删除前先释放算力节点上的模型
+    await _unload_from_compute_node()
     get_platform_store().delete_compare_task(task_id)
     return ok({"deleted": task_id})
 
@@ -1053,26 +1128,56 @@ async def model_compare_update_load_status(task_id: str, payload: dict[str, Any]
 
 @router.post("/model-compare/{task_id}/load")
 async def model_compare_load(task_id: str) -> dict[str, Any]:
+    """真正加载模型到算力节点（不再使用假 PID/端口）。"""
     try:
-        task = get_platform_store().compare_task(task_id)
+        store = get_platform_store()
+        task = store.compare_task(task_id)
         models = task.get("models") or []
         if isinstance(models, str):
             try:
                 models = json.loads(models)
             except json.JSONDecodeError:
                 models = []
-        loaded_models = [
-            {
-                "model_id": item.get("model_id"),
-                "model_name": item.get("model_name"),
-                "status": "ready",
-                "pid": 45000 + index,
-                "port": item.get("port") or 18000 + index,
+        # 选取在线算力节点
+        node = _select_first_online_node(store)
+        if not node:
+            return ok({"status": "failed", "error": "no online compute node"})
+        client = ComputeNodeClient(node["api_base_url"])
+        loaded_models = []
+        for item in models:
+            if not isinstance(item, dict):
+                continue
+            model_path = item.get("model_path", "")
+            if not model_path:
+                # 尝试从模型库获取路径
+                model_id = item.get("model_id", "")
+                try:
+                    db_model = store.model(model_id)
+                    model_path = db_model.get("path", "")
+                except KeyError:
+                    pass
+            if not model_path:
+                loaded_models.append({**item, "status": "error", "error": "model_path not found"})
+                continue
+            # 真正调用算力节点加载模型
+            load_payload = {
+                "model_name_or_path": model_path,
+                "template": item.get("template", "qwen"),
             }
-            for index, item in enumerate(models)
-            if isinstance(item, dict)
-        ]
-        return ok(get_platform_store().update_compare_task(task_id, {"status": "loaded", "load_status": {"loaded_models": loaded_models}}))
+            if item.get("adapter_path"):
+                load_payload["adapter_name_or_path"] = item["adapter_path"]
+            try:
+                result = await client._request("POST", "/inference/load", json_data=load_payload)
+                if result.get("loaded"):
+                    store.mark_inference_loaded(node["id"])
+                    loaded_models.append({**item, "status": "ready"})
+                else:
+                    loaded_models.append({**item, "status": "error", "error": result.get("error", "load failed")})
+            except Exception as exc:
+                loaded_models.append({**item, "status": "error", "error": str(exc)})
+        status = "loaded" if any(m.get("status") == "ready" for m in loaded_models) else "failed"
+        updated = store.update_compare_task(task_id, {"status": status, "load_status": {"loaded_models": loaded_models}})
+        return ok(updated)
     except KeyError:
         raise fail(404, "compare task not found")
 
@@ -1080,7 +1185,11 @@ async def model_compare_load(task_id: str) -> dict[str, Any]:
 @router.post("/model-compare/{task_id}/unload")
 async def model_compare_unload(task_id: str) -> dict[str, Any]:
     try:
-        return ok(get_platform_store().update_compare_task(task_id, {"status": "pending", "load_status": {"loaded_models": []}}))
+        store = get_platform_store()
+        # 真正释放算力节点上的模型资源
+        unload_result = await _unload_from_compute_node()
+        updated = store.update_compare_task(task_id, {"status": "pending", "load_status": {"loaded_models": []}})
+        return ok({"task": updated, "unload": unload_result})
     except KeyError:
         raise fail(404, "compare task not found")
 
@@ -1092,18 +1201,23 @@ async def model_compare_start_model(task_id: str, payload: dict[str, Any] = Body
 
 @router.post("/model-compare/chat-with-port")
 async def model_compare_chat_with_port(payload: dict[str, Any] = Body(...)) -> dict[str, Any]:
-    question = ""
-    for message in payload.get("messages") or []:
-        if message.get("role") == "user":
-            question = str(message.get("content") or "")
-    content = f"当前后端已收到推理请求：{question[:120]}"
-    return ok({"response": content, "content": content})
+    """Proxy non-streaming chat to the compute node running the inference model."""
+    store = get_platform_store()
+    node = _select_first_online_node(store)
+    if not node:
+        return ok({"response": "no online compute node available for inference", "request": payload})
+    try:
+        client = ComputeNodeClient(node["api_base_url"])
+        result = await client._request("POST", "/inference/chat", json_data=_build_messages_payload(payload))
+        return ok(result)
+    except Exception as exc:
+        return ok({"response": f"inference failed: {exc}", "request": payload})
 
 
 @router.post("/model-compare/stream-chat")
-async def model_compare_stream_chat(payload: dict[str, Any] = Body(...)) -> dict[str, Any]:
-    question = payload.get("user_question") or payload.get("question") or ""
-    return ok({"response": f"当前后端已收到流式推理请求：{str(question)[:120]}"})
+async def model_compare_stream_chat(payload: dict[str, Any] = Body(...)) -> StreamingResponse:
+    """Stream chat from the compute node (SSE proxy)."""
+    return await _stream_chat_proxy(payload)
 
 
 @router.post("/model-chat/batch")
@@ -1120,7 +1234,7 @@ async def model_chat_local(payload: dict[str, Any] = Body(...)) -> dict[str, Any
         return ok({"response": "no online compute node available for inference", "request": payload})
     try:
         client = ComputeNodeClient(node["api_base_url"])
-        result = await client._request("POST", "/inference/chat", json_data=payload)
+        result = await client._request("POST", "/inference/chat", json_data=_build_messages_payload(payload))
         return ok(result)
     except Exception as exc:
         return ok({"response": f"inference failed: {exc}", "request": payload})
@@ -1129,28 +1243,15 @@ async def model_chat_local(payload: dict[str, Any] = Body(...)) -> dict[str, Any
 @router.post("/model-chat/local/chat/stream")
 async def model_chat_local_stream(payload: dict[str, Any] = Body(...)) -> StreamingResponse:
     """Stream chat from the compute node."""
-    store = get_platform_store()
-    node = _select_first_online_node(store)
-    if not node:
-        return StreamingResponse(
-            iter(['data: {"error": "no online compute node"}\n\n']),
-            media_type="text/event-stream",
-        )
-    client = ComputeNodeClient(node["api_base_url"])
-
-    async def stream_proxy():
-        async with httpx.AsyncClient(timeout=300) as http:
-            url = f"{node['api_base_url'].rstrip('/')}/modelTF/inference/chat/stream"
-            async with http.stream("POST", url, json=payload, headers=client.headers()) as resp:
-                async for chunk in resp.aiter_bytes():
-                    yield chunk
-
-    return StreamingResponse(stream_proxy(), media_type="text/event-stream")
+    return await _stream_chat_proxy(payload)
 
 
 @router.post("/model-chat/local/preload")
 async def model_chat_local_preload(payload: dict[str, Any] = Body(...)) -> dict[str, Any]:
     """Load a model on the compute node for inference."""
+    model_path = (payload.get("model_name_or_path") or "").strip()
+    if not model_path:
+        return ok({"loaded": False, "error": "model_name_or_path is required"})
     store = get_platform_store()
     node = _select_first_online_node(store)
     if not node:
@@ -1158,6 +1259,8 @@ async def model_chat_local_preload(payload: dict[str, Any] = Body(...)) -> dict[
     try:
         client = ComputeNodeClient(node["api_base_url"])
         result = await client._request("POST", "/inference/load", json_data=payload)
+        if result.get("loaded"):
+            store.mark_inference_loaded(node["id"])
         return ok(result)
     except Exception as exc:
         return ok({"loaded": False, "error": str(exc)})
@@ -1167,6 +1270,9 @@ async def model_chat_local_preload(payload: dict[str, Any] = Body(...)) -> dict[
 async def model_chat_local_unload() -> dict[str, Any]:
     """Unload the inference model from the compute node."""
     store = get_platform_store()
+    # Clear all inference tracking
+    for n in store.compute_nodes():
+        store.mark_inference_unloaded(n["id"])
     node = _select_first_online_node(store)
     if not node:
         return ok({"unloaded": False, "error": "no online compute node"})
@@ -1196,6 +1302,9 @@ async def model_chat_local_status() -> dict[str, Any]:
 @router.post("/model-chat/trained/preload")
 async def model_chat_trained_preload(payload: dict[str, Any] = Body(...)) -> dict[str, Any]:
     """Load a trained model (base + adapter) on the compute node for inference."""
+    model_path = (payload.get("model_name_or_path") or "").strip()
+    if not model_path:
+        return ok({"loaded": False, "error": "model_name_or_path is required"})
     store = get_platform_store()
     node = _select_first_online_node(store)
     if not node:
@@ -1203,6 +1312,8 @@ async def model_chat_trained_preload(payload: dict[str, Any] = Body(...)) -> dic
     try:
         client = ComputeNodeClient(node["api_base_url"])
         result = await client._request("POST", "/inference/load", json_data=payload)
+        if result.get("loaded"):
+            store.mark_inference_loaded(node["id"])
         return ok(result)
     except Exception as exc:
         return ok({"loaded": False, "error": str(exc)})

backend/app/db/platform_store.py

@@ -293,6 +293,19 @@ class PlatformStore:
         self.database_url = _psycopg_url(database_url or settings.database_url)
         self.ensure_schema()
         self.ensure_seed_data()
+        # Track which compute nodes have an active inference model loaded
+        self._inference_nodes: set[str] = set()
+
+    # ── inference node tracking ────────────────────────────────────
+
+    def mark_inference_loaded(self, node_id: str) -> None:
+        self._inference_nodes.add(node_id)
+
+    def mark_inference_unloaded(self, node_id: str) -> None:
+        self._inference_nodes.discard(node_id)
+
+    def is_inference_loaded(self, node_id: str) -> bool:
+        return node_id in self._inference_nodes
 
     @contextmanager
     def connect(self) -> Iterator["PgConnection"]:
@@ -2698,6 +2711,11 @@ class PlatformStore:
                 )
                 busy = task is not None and task.get("status") == "running"
                 reserved = task is not None and task.get("status") in {"syncing", "queued"}
+                # Also mark GPU as busy if an inference model is loaded on this node
+                inference_busy = self.is_inference_loaded(row["node_id"])
+                if inference_busy and not busy:
+                    busy = True
+                    reserved = False
                 memory_used = round(row["memory_total_gb"] * (0.72 if busy else 0.18 if reserved else 0.04), 1)
                 gpu_percent = 86 if busy else 22 if reserved else 3
                 memory_total = float(row["memory_total_gb"] or 0)