Initial commit

Dockerfile

@@ -0,0 +1,14 @@
+FROM modelhubxc-4pd.tencentcloudcr.com/xc_agent_platform/python:3.11-slim
+
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt
+
+COPY . .
+
+EXPOSE 8080
+
+CMD ["python", "main.py"]

main.py

@@ -0,0 +1,155 @@
+import requests
+import json
+from typing import List, Tuple
+
+# ========== 全局配置 ==========
+BASE_URL = "https://modelhub.org.cn"
+LOGIN_ENDPOINT = "/adminApi/user/login"
+SUBMIT_TEST_TASK_ENDPOINT = "/adminApi/async/task/create-contest-task"
+
+USER_ACCOUNT = "zhoushasha@4paradigm.com"
+USER_PASSWORD = "4pdpassword"
+
+
+CONTEST_API_TOKEN = "ef1ef82f3c9efee413d602345fbe224d"
+
+
+CONTRIBUTORS = "zhoushasha"
+
+
+GPU_TYPE = "Cambricon_mlu-370-x8"
+
+TASK_TYPE = "text-generation"
+
+HEADERS = {"Content-Type": "application/json"}
+
+# ======== 模型列表（保持不变）========
+ALL_MODEL_IDS = [
+    
+    "AI-ModelScope/gemma-2b",
+    "AI-ModelScope/falcon-mamba-7b",
+    "katanemo/deepseek-2",
+    "OpenBMB/MiniCPM4-0.5B",
+    "NousResearch/Meta-Llama-3-8B-Instruct",
+    "MediaTek-Research/Breeze-7B-Instruct-v1_0",
+    "QLUNLP/BianCang-Qwen2.5-7B-Instruct",
+    "OpenBMB/MiniCPM4-Survey",
+    "OpenBMB/MiniCPM4-8B",
+    "PaddlePaddle/ERNIE-4.5-0.3B-PT",
+    "LLM-Research/Llama-Guard-3-8B",
+    "OpenBMB/MiniCPM-2B-dpo-fp16",
+    "OpenBMB/MiniCPM4.1-8B",
+    "Cylingo/Xinyuan-LLM-14B-0428",
+    "Fengshenbang/Ziya-LLaMA-13B-v1",
+    "baichuan-inc/Baichuan2-13B-Chat",
+    "LLM-Research/gemma-2-9b-it",
+    "Qwen/CodeQwen1.5-7B-Chat",
+    "OpenBMB/cpm-bee-10b",
+    "OpenBMB/MiniCPM3-4B",
+    
+
+]
+
+# === 登录获取 token ===
+def login():
+    payload = {"userAccount": USER_ACCOUNT, "userPassword": USER_PASSWORD}
+    print("🔑 正在登录...")
+    resp = requests.post(BASE_URL + LOGIN_ENDPOINT, headers=HEADERS, json=payload)
+    if resp.status_code != 200:
+        raise Exception(f"HTTP 登录失败: {resp.text}")
+    data = resp.json()
+    if data.get("code") != 0:
+        raise Exception(f"业务登录失败: {data.get('message')}")
+    token = data["data"]["token"]
+    print("✅ 登录成功！")
+    return token
+
+# === 提交单个模型的测试任务（vLLM + kunlunxin_p-800）===
+def submit_test_task(token: str, model_id: str) -> Tuple[str, str]:
+    auth_headers = {**HEADERS, "Authorization": f"Bearer {token}"}
+    config_content = f"""docker_image: harbor.4pd.io/hardcore-tech/cambricon-mlu370-pytorch:v25.01-torch2.5.0-torchmlu1.24.1-ubuntu22.04-py310
+nv_docker_image: harbor.4pd.io/dooke/vllm/vllm/vllm-openai:v0.11.0
+framework: vllm
+storage: gpfs
+modelhub_options:
+  srcRelativePath: leaderboard/modelHubXC/{model_id}
+  mountPoint: /model
+sut_config:
+  values:
+    gpu_num: 1
+    env:
+      - name: MAX_MODEL_LEN
+        value: 8192
+    command: ["vllm", "serve", "/model", "--port", "8000", "--served-model-name", "llm", "--max-model-len", "8192", "--trust-remote-code", "--dtype", "float16"]
+ref_config:
+  values:
+    cpu_num: 2
+    gpu_num: 1
+    env:
+      - name: MAX_MODEL_LEN
+        value: 8192
+    command: ["vllm", "serve", "/model", "--port", "80", "--served-model-name", "llm", "--max-model-len", "8192", "--trust-remote-code", "--dtype", "float16"]
+"""
+    task_data = {
+        "contestApiToken": CONTEST_API_TOKEN,
+        "contributors": CONTRIBUTORS,
+        "gpuTypes": [GPU_TYPE],
+        "taskType": TASK_TYPE,
+        "modelId": model_id,
+        "framework": "vllm",
+        "submissionConfig": [{
+            "config": config_content,
+            "gpuType": GPU_TYPE,
+            "taskType": TASK_TYPE
+        }]
+    }
+    print(f"📤 提交验证任务: {model_id} (GPU: {GPU_TYPE})")
+    try:
+        resp = requests.post(BASE_URL + SUBMIT_TEST_TASK_ENDPOINT, json=task_data, headers=auth_headers, timeout=15)
+        if resp.status_code == 200:
+            result = resp.json()
+            if result.get("code") == 0:
+                task_id = result.get("data", {}).get("id")
+                print(f"✅ 验证任务提交成功! Task ID: {task_id}")
+                return task_id, model_id
+            else:
+                print(f"❌ 验证任务业务错误 ({model_id}): {result.get('message')}")
+                return None, model_id
+        else:
+            print(f"❌ 验证任务 HTTP 错误 ({model_id}): {resp.status_code} - {resp.text}")
+            return None, model_id
+    except Exception as e:
+        print(f"💥 提交验证任务异常 ({model_id}): {e}")
+        return None, model_id
+
+# === 主函数：仅提交验证任务 ===
+def main():
+    if not ALL_MODEL_IDS:
+        print("❌ 模型列表为空，请在 ALL_MODEL_IDS 中填入模型ID")
+        return
+
+    token = login()
+    total_count = len(ALL_MODEL_IDS)
+    print(f"📊 共 {total_count} 个模型待提交验证任务")
+
+    successful_tasks: List[Tuple[str, str]] = []  # (task_id, model_id)
+
+    for model_id in ALL_MODEL_IDS:
+        task_id, mid = submit_test_task(token, model_id)
+        if task_id:
+            successful_tasks.append((task_id, mid))
+
+    # 写入成功提交的 task_id 和 model_id 到文件
+    with open("submitted_validation_tasks.txt", "w", encoding="utf-8") as f:
+        for tid, mid in successful_tasks:
+            f.write(f"{tid}\t{mid}\n")
+
+    # 最终统计
+    print("\n" + "=" * 60)
+    print(f"🎉 全部完成！")
+    print(f"✅ 成功提交验证任务: {len(successful_tasks)}")
+    print(f"📄 详情已写入: submitted_validation_tasks.txt")
+    print(f"📊 总计尝试: {total_count}")
+
+if __name__ == "__main__":
+    main()