{
  "id": "tensorflow/triton-inference-version-mismatch",
  "signature": "tensorflow.python.framework.errors_impl.InternalError: TRITONBACKEND_ModelInstanceInitialize: model 'resnet50' version 1 has unsupported TensorFlow runtime version. Expected 2.12.0, got 2.10.0",
  "signature_zh": "tensorflow.python.framework.errors_impl.InternalError: TRITONBACKEND_ModelInstanceInitialize: 模型 'resnet50' 版本1的 TensorFlow 运行时版本不受支持。期望 2.12.0，实际为 2.10.0",
  "regex": "TRITONBACKEND_ModelInstanceInitialize.*unsupported TensorFlow runtime version",
  "domain": "tensorflow",
  "category": "config_error",
  "subcategory": null,
  "root_cause": "The TensorFlow runtime version compiled into the Triton Inference Server backend does not match the version required by the saved model.",
  "root_cause_type": "generic",
  "root_cause_zh": "Triton 推理服务器后端编译的 TensorFlow 运行时版本与保存模型所需的版本不匹配。",
  "versions": [
    {
      "version": "2.10",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "2.12",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    },
    {
      "version": "2.13",
      "introduced": null,
      "deprecated": null,
      "removed": null,
      "behavior_change": null,
      "status": "active"
    }
  ],
  "os_specific": {},
  "dead_ends": [
    {
      "action": "Reinstall Triton Inference Server with the latest version",
      "why_fails": "The latest Triton may still bundle a different TensorFlow version; you need to match the exact version.",
      "fail_rate": 0.5,
      "condition": "",
      "sources": []
    },
    {
      "action": "Convert the model to ONNX format",
      "why_fails": "ONNX conversion may not support all TF ops, and the error is about runtime version, not model format.",
      "fail_rate": 0.6,
      "condition": "",
      "sources": []
    }
  ],
  "workarounds": [
    {
      "action": "Use a Triton Inference Server Docker image that matches the TensorFlow version of your model. Check the tag at https://catalog.ngc.nvidia.com/orgs/nvidia/containers/tritonserver and pull the correct version, e.g., nvcr.io/nvidia/tritonserver:23.12-tf2-py3 for TF 2.12.",
      "success_rate": 0.9,
      "how": "Use a Triton Inference Server Docker image that matches the TensorFlow version of your model. Check the tag at https://catalog.ngc.nvidia.com/orgs/nvidia/containers/tritonserver and pull the correct version, e.g., nvcr.io/nvidia/tritonserver:23.12-tf2-py3 for TF 2.12.",
      "condition": "",
      "sources": []
    },
    {
      "action": "Export the model using the same TensorFlow version as the Triton backend. Run 'pip install tensorflow==2.12.0' and re-save the model with tf.saved_model.save().",
      "success_rate": 0.85,
      "how": "Export the model using the same TensorFlow version as the Triton backend. Run 'pip install tensorflow==2.12.0' and re-save the model with tf.saved_model.save().",
      "condition": "",
      "sources": []
    }
  ],
  "workarounds_zh": [
    "Use a Triton Inference Server Docker image that matches the TensorFlow version of your model. Check the tag at https://catalog.ngc.nvidia.com/orgs/nvidia/containers/tritonserver and pull the correct version, e.g., nvcr.io/nvidia/tritonserver:23.12-tf2-py3 for TF 2.12.",
    "Export the model using the same TensorFlow version as the Triton backend. Run 'pip install tensorflow==2.12.0' and re-save the model with tf.saved_model.save()."
  ],
  "transition_graph": {
    "leads_to": [],
    "preceded_by": [],
    "frequently_confused_with": []
  },
  "official_doc_url": "https://docs.nvidia.com/deeplearning/triton-inference-server/user-guide/docs/user_guide/model_configuration.html",
  "official_doc_section": null,
  "error_code": "ETRV",
  "verification_tier": "ai_generated",
  "confidence": 0.85,
  "fix_success_rate": 0.88,
  "resolvable": "true",
  "first_seen": "2024-06-20",
  "last_confirmed": "2024-06-01",
  "last_updated": "2024-06-01",
  "evidence_count": 1,
  "tags": [],
  "locale": "en",
  "aliases": []
}