> For the complete documentation index, see [llms.txt](https://klaralabs.gitbook.io/klara-labs-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://klaralabs.gitbook.io/klara-labs-docs/technology/benchmark-and-performances.md). # Benchmark & Performances Our hallucination detection model achieves state-of-the-art performance while maintaining exceptional inference speed. The architecture leverages advanced transformer techniques to deliver accurate results with minimal computational requirements. ### Performance Highlights * **80.3% F1 Score** on the RAGTruth benchmark, outperforming previous encoder-based models by a significant margin ### Efficiency Metrics * **30 example/s processed** on a single GPU * **Compact model size** (300M param.) * **30x smaller** than comparable LLM-based detection systems * **Long context support** up to 4,096 tokens ### Real-world Applications The model's combination of high accuracy and processing speed makes it ideal for: * **Real-time content moderation** systems requiring immediate hallucination detection * **High-volume document processing** workflows with RAG components * **Resource-constrained environments** where deploying large LLMs isn't feasible Our benchmarks demonstrate that lightweight, purpose-built models can achieve excellent hallucination detection performance while maintaining the speed and efficiency needed for production environments. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://klaralabs.gitbook.io/klara-labs-docs/technology/benchmark-and-performances.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.