Skip to main content
In many real-world scenarios, you may want to use multiple Retriever or Generation modules within the same Pipeline to perform different logical tasks, such as hybrid retrieval or multi-agent systems.
In fact, this can be achieved simply by configuring different parameters for the same module. To support this, UR-2.0 provides a simple and flexible mechanism —
you can reuse and independently invoke the same Server module by assigning different aliases to it.

​
Usage Example

​
Step 1: Configure Aliased Servers

In the pipeline.yaml file, you can define multiple aliases for the same module path under the servers field:
https://mintcdn.com/ultrarag/T7GffHzZitf6TThi/images/yaml.svg?fit=max&auto=format&n=T7GffHzZitf6TThi&q=85&s=69b41e79144bc908039c2ee3abbb1c3bexamples/hybrid_search.yaml
# MCP Server
servers:
  benchmark: servers/benchmark
  dense: servers/retriever
  bm25: servers/retriever
  custom: servers/custom

# MCP Client Pipeline
pipeline:
- benchmark.get_data
- dense.retriever_init
- bm25.retriever_init
- dense.retriever_search:
    output:
      ret_psg: dense_psg
- bm25.bm25_search:
    output:
      ret_psg: sparse_psg
- custom.merge_passages:
    input:
      ret_psg: dense_psg
      temp_psg: sparse_psg
In this example, both dense and bm25 point to the same module path servers/retriever,
but they are built and invoked as two independent Server instances.

​
Step 2: Call Them Separately in the Pipeline

In the Pipeline definition, you can use these aliases just like separate modules:
https://mintcdn.com/ultrarag/T7GffHzZitf6TThi/images/yaml.svg?fit=max&auto=format&n=T7GffHzZitf6TThi&q=85&s=69b41e79144bc908039c2ee3abbb1c3bexamples/hybrid_search.yaml
# MCP Server
servers:
  benchmark: servers/benchmark
  dense: servers/retriever
  bm25: servers/retriever
  custom: servers/custom

# MCP Client Pipeline
pipeline:
- benchmark.get_data
- dense.retriever_init
- bm25.retriever_init
- dense.retriever_search:
    output:
      ret_psg: dense_psg
- bm25.bm25_search:
    output:
      ret_psg: sparse_psg
- custom.merge_passages:
    input:
      ret_psg: dense_psg
      temp_psg: sparse_psg
At runtime, UltraRAG will automatically distinguish between these two instances:
each alias corresponds to an independent parameter file, runtime context, and cache space —
enabling multiple modules to run in parallel without interfering with one another.