LlamaCppChatGenerator

Complete chats using large language models running on llama.cpp.

Key Features

• Runs LLMs locally using llama.cpp, enabling inference without GPUs on standard hardware.
• Uses the quantized GGUF format, which reduces memory requirements and speeds up inference.
• Supports tool calling for agentic workflows.
• Configurable context window size and batch size for memory management.
• Compatible with any GGUF model available on Hugging Face.

Configuration

1. Drag the LlamaCppChatGenerator component onto the canvas from the Component Library.
2. Click the component to open the configuration panel.
3. On the General tab:
   1. Enter the local file path to the GGUF model file, for example /Downloads/gemma-3-1b-it-q4_0.gguf.
4. Go to the Advanced tab to configure the context size (n_ctx), batch size (n_batch), model kwargs, and generation kwargs.

Connections

LlamaCppChatGenerator receives a messages list of ChatMessage objects from ChatPromptBuilder. It outputs replies (a list of ChatMessage objects). Connect its replies output to OutputAdapter before passing to AnswerBuilder or DeepsetAnswerBuilder.

Usage Example

This is an example RAG pipeline with LlamaCppChatGenerator and DeepsetAnswerBuilder connected through OutputAdapter:

components:
  bm25_retriever: # Selects the most similar documents from the document store
    type: haystack_integrations.components.retrievers.opensearch.bm25_retriever.OpenSearchBM25Retriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          hosts:
          index: 'Standard-Index-English'
          max_chunk_bytes: 104857600
          embedding_dim: 768
          return_embedding: false
          method:
          mappings:
          settings:
          create_index: true
          http_auth:
          use_ssl:
          verify_certs:
          timeout:
      top_k: 20 # The number of results to return
      fuzziness: 0

  query_embedder:
    type: deepset_cloud_custom_nodes.embedders.nvidia.text_embedder.DeepsetNvidiaTextEmbedder
    init_parameters:
      normalize_embeddings: true
      model: intfloat/e5-base-v2

  embedding_retriever: # Selects the most similar documents from the document store
    type: haystack_integrations.components.retrievers.opensearch.embedding_retriever.OpenSearchEmbeddingRetriever
    init_parameters:
      document_store:
        type: haystack_integrations.document_stores.opensearch.document_store.OpenSearchDocumentStore
        init_parameters:
          hosts:
          index: 'Standard-Index-English'
          max_chunk_bytes: 104857600
          embedding_dim: 768
          return_embedding: false
          method:
          mappings:
          settings:
          create_index: true
          http_auth:
          use_ssl:
          verify_certs:
          timeout:
      top_k: 20 # The number of results to return

  document_joiner:
    type: haystack.components.joiners.document_joiner.DocumentJoiner
    init_parameters:
      join_mode: concatenate

  ranker:
    type: deepset_cloud_custom_nodes.rankers.nvidia.ranker.DeepsetNvidiaRanker
    init_parameters:
      model: intfloat/simlm-msmarco-reranker
      top_k: 8

  meta_field_grouping_ranker:
    type: haystack.components.rankers.meta_field_grouping_ranker.MetaFieldGroupingRanker
    init_parameters:
      group_by: file_id
      subgroup_by:
      sort_docs_by: split_id

  answer_builder:
    type: deepset_cloud_custom_nodes.augmenters.deepset_answer_builder.DeepsetAnswerBuilder
    init_parameters:
      reference_pattern: acm

  ChatPromptBuilder:
    type: haystack.components.builders.chat_prompt_builder.ChatPromptBuilder
    init_parameters:
      template:
      - _content:
        - text: "You are a helpful assistant answering the user's questions based on the provided documents.\nIf the answer is not in the documents, rely on the web_search tool to find information.\nDo not use your own knowledge.\n"
        _role: system
      - _content:
        - text: "Provided documents:\n{% for document in documents %}\nDocument [{{ loop.index }}] :\n{{ document.content }}\n{% endfor %}\n\nQuestion: {{ query }}\n"
        _role: user
      required_variables:
      variables:
  OutputAdapter:
    type: haystack.components.converters.output_adapter.OutputAdapter
    init_parameters:
      template: '{{ replies[0] }}'
      output_type: List[str]
      custom_filters:
      unsafe: false

  LlamaCppChatGenerator:
    type: haystack_integrations.components.generators.llama_cpp.chat.chat_generator.LlamaCppChatGenerator
    init_parameters:
      model: /Downloads/gemma-3-1b-it-q4_0.gguf
      n_ctx: 0
      n_batch: 512
      model_kwargs:
      generation_kwargs:

connections:  # Defines how the components are connected
- sender: bm25_retriever.documents
  receiver: document_joiner.documents
- sender: query_embedder.embedding
  receiver: embedding_retriever.query_embedding
- sender: embedding_retriever.documents
  receiver: document_joiner.documents
- sender: document_joiner.documents
  receiver: ranker.documents
- sender: ranker.documents
  receiver: meta_field_grouping_ranker.documents
- sender: meta_field_grouping_ranker.documents
  receiver: answer_builder.documents
- sender: meta_field_grouping_ranker.documents
  receiver: ChatPromptBuilder.documents
- sender: OutputAdapter.output
  receiver: answer_builder.replies
- sender: ChatPromptBuilder.prompt
  receiver: LlamaCppChatGenerator.messages
- sender: LlamaCppChatGenerator.replies
  receiver: OutputAdapter.replies

inputs:  # Define the inputs for your pipeline
  query:  # These components will receive the query as input
  - "bm25_retriever.query"
  - "query_embedder.text"
  - "ranker.query"
  - "answer_builder.query"
  - "ChatPromptBuilder.query"
  filters:  # These components will receive a potential query filter as input
  - "bm25_retriever.filters"
  - "embedding_retriever.filters"

outputs:  # Defines the output of your pipeline
  documents: "meta_field_grouping_ranker.documents"  # The output of the pipeline is the retrieved documents
  answers: "answer_builder.answers"  # The output of the pipeline is the generated answers

max_runs_per_component: 100

metadata: {}

Parameters

Inputs

Parameter	Type	Default	Description
messages	List[ChatMessage]		A list of ChatMessage objects representing the input messages.
generation_kwargs	Optional[Dict[str, Any]]	None	A dictionary containing keyword arguments to customize text generation. For more information on the arguments you can use, see llama.cpp documentation.
tools	Optional[Union[List[Tool], Toolset]]	None	A list of tools or a Toolset for which the model can prepare calls.

Outputs

Parameter	Type	Default	Description
replies	List[ChatMessage]		The responses from the model.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
model	str		The path of a quantized model for text generation, for example, "zephyr-7b-beta.Q4_0.gguf". If the model path is also specified in the model_kwargs, this parameter is ignored.
n_ctx	Optional[int]	0	The number of tokens in the context. When set to 0, the context is taken from the model.
n_batch	Optional[int]	512	Prompt processing maximum batch size.
model_kwargs	Optional[Dict[str, Any]]	None	Dictionary containing keyword arguments used to initialize the LLM for text generation. These keyword arguments provide fine-grained control over the model loading. In case of duplication, these kwargs override model, n_ctx, and n_batch init parameters. For more information on the available kwargs, see llama.cpp documentation.
generation_kwargs	Optional[Dict[str, Any]]	None	A dictionary containing keyword arguments to customize text generation. For more information on the available kwargs, see llama.cpp documentation.
tools	Optional[Union[List[Tool], Toolset]]	None	A list of tools or a Toolset for which the model can prepare calls. This parameter can accept either a list of Tool objects or a Toolset instance.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
messages	List[ChatMessage]		A list of ChatMessage instances representing the input messages.
generation_kwargs	Optional[Dict[str, Any]]	None	A dictionary containing keyword arguments to customize text generation. For more information on the available kwargs, see llama.cpp documentation.
tools	Optional[Union[List[Tool], Toolset]]	None	A list of tools or a Toolset for which the model can prepare calls. If set, it will override the tools parameter set in pipeline configuration.