export PATH=$PATH:~/.cargo/bin/
# 设置TEI运行计算卡id号与模型后端 
export TEI_NPU_DEVICE=0  # 按需设置计算卡id
export TEI_NPU_BACKEND=mindietorch  # 按需选择mindietorch或atb
# 本地模型权重路径或在Huggingface代码仓中的模型id
model_path_reranker=/workspace/cache/bge-reranker-v2-m3

# 以下启动方式及参数名与原生TEI一致，请按需选择拉起Embedding或Reranker模型
# Reranker模型
text-embeddings-router --model-id $model_path_reranker --dtype float16 --max-client-batch-size 192 --max-concurrent-requests 2048 --max-batch-tokens 163840 --max-batch-requests 128 --port 8080

用开发实例运行脚本，输出：
[root@pod-13614897552 ~]# ./start.sh 
2025-04-15T02:29:37.931245Z  INFO text_embeddings_router: router/src/main.rs:140: Args { model_id: "/wor******/*****/***-********-*2-m3", revision: None, tokenization_workers: None, dtype: Some(Float16), pooling: None, max_concurrent_requests: 2048, max_batch_tokens: 163840, max_batch_requests: Some(128), max_client_batch_size: 192, auto_truncate: false, hf_api_token: None, hostname: "pod-13614897552 ", port: 8080, uds_path: "/tmp/text-embeddings-inference-server", huggingface_hub_cache: None, payload_limit: 2000000, api_key: None, json_output: false, otlp_endpoint: None, cors_allow_origin: None }
2025-04-15T02:29:38.594544Z  WARN text_embeddings_router: router/src/lib.rs:165: Could not find a Sentence Transformers config
2025-04-15T02:29:38.594573Z  INFO text_embeddings_router: router/src/lib.rs:169: Maximum number of tokens per request: 8192
2025-04-15T02:29:38.596265Z  INFO text_embeddings_core::tokenization: core/src/tokenization.rs:23: Starting 12 tokenization workers
2025-04-15T02:29:40.345473Z  INFO text_embeddings_router: router/src/lib.rs:194: Starting model backend
2025-04-15T02:29:40.345971Z  INFO text_embeddings_backend_python::management: backends/python/src/management.rs:54: Starting Python backend
Error: Could not create backend

Caused by:
    Could not start backend: Python backend failed to start