# Vectara's models

Vectara's advanced end-to-end RAG models have pioneered enterprise-grade RAG.

- **Boomerang**  
  Retrieval model  
  * * *  
  [Learn more](https://www.vectara.com/business/platform/models#boomerang)  
- **Mockingbird**  
  Generation model  
  * * *  
  [Learn more](https://www.vectara.com/business/platform/models#mockingbird)  
- **Slingshot**  
  Multilingual reranker  
  * * *  
  [Learn more](https://www.vectara.com/business/platform/models#slingshot)  
- **Hughes HEM**  
  Evaluation model  
  * * *  
  [Learn more](https://www.vectara.com/business/platform/models#hughes-hallucination-evaluation-model-hhem)

**Best-of-breed models at every stage**

A RAG pipeline is only as good as the models that power it.

### Multilingual semantic search

[See announcement](https://www.vectara.com/blog/introducing-boomerang-vectaras-new-and-improved-retrieval-model)

Boomerang manages retrieval in RAG, grounding results in datasets and supporting Hybrid Search by mapping concepts to vectors, not keywords.

Boomerang supports hundreds of languages and dialects and uses zero-shot learning to quickly find relevant information across diverse domains.

### Boomerang’s performance benchmarks

[Performance details (June 2024)](https://www.vectara.com/blog/the-latest-benchmark-between-vectara-openai-and-coheres-embedding-models)

- ### Multi-domain question and answer

Boomerang is on par with OpenAI and Cohere embedding models.  
- ### Cross-lingual retrieval

Boomerang outperforms all other leading models on cross-lingual retrieval.  
- ### Quality vs embedding size

Boomerang optimizes cost by balancing high quality and small embedding size.

### Reranking for relevance and variety

[See documentation](https://docs.vectara.com/docs/api-reference/search-apis/reranking)

The Slingshot model reorganizes search results to ensure relevance and avoid redundancy, delivering accurate outcomes across queries.

It can be combined with inputs based on your priorities, such as recency or keyword frequency, to ensure the most relevant information rises to the top.

### Slingshot’s performance benchmarks

[Performance details (May 2024)](https://www.vectara.com/blog/deep-dive-into-vectara-multilingual-reranker-v1-state-of-the-art-reranker-across-100-languages)

- ### Reranker English-language performance

Slingshot outperforms Cohere and other major models on English-language rerank.  
- ### Reranker multilingual performance

Slingshot outperforms most major models on multi-language rerank.

### Context-aware multilingual generation

[Explore documentation](https://docs.vectara.com/docs/learn/mockingbird-llm)

The Mockingbird LLM is the “G” in our RAG platform. We’ve optimized it to deliver highly relevant answers and cite sources for each claim.

Mockingbird handles queries and results from diverse domains and languages. It can generate everything modern AI applications need.

### Mockingbird’s performance benchmarks

[Performance details (July 2024)](https://www.vectara.com/blog/mockingbird-a-rag-and-structured-output-focused-llm)

- ### English RAG output quality

Mockingbird outperforms GPT4 and other major models in English-language output quality.  
- ### Multilingual RAG output quality

Mockingbird outperforms GPT4 and other major models in multilingual output quality.  
- ### Structured output accuracy

Mockingbird outperforms GPT4 and other major models in structured output accuracy.

### Hallucination detection for any LLM

[View documentation](https://docs.vectara.com/docs/learn/hallucination-evaluation)

The Hughes Hallucination Evaluation Model assesses LLMs for hallucinated results versus factual consistency.

The free version is on Hugging Face, with a more advanced version used in our platform's Factual Consistency Score.

## HHEM’s hallucination detection performance benchmarks

[Performance details (August 2024)](https://www.vectara.com/blog/hhem-2-1-a-better-hallucination-detection-model/)

## Remembering a champion of responsible AI

The Hughes Hallucination Evaluation Model honors Dr. Simon Mark Hughes, a pioneering machine learning engineer at Vectara. His research on AI hallucination detection, before his passing in 2023, continues to shape industry efforts toward truthful and reliable AI outputs.

## Create AI Agents in minutes

Vectara is the Agent OS for trusted enterprise AI: governed, grounded, auditable agents with real-time policy and factual-consistency enforcement.