Wrenai is an open source generation business intelligence (GenBI) agent developed by Canner to enable seamless, natural language to interact with structured data. It targets technical and non-technical teams and provides tools to query, analyze and visualize data without writing SQL. All features and integrations will be verified in accordance with official documentation and the latest version.
Key Features
- SQL’s natural language:
Users can ask data questions in normal languages (cross multiple languages) and convert them into accurate production-grade SQL queries. This simplifies data access to non-technical users. - Multi-mode output:
The platform generates SQL, charts, summary reports, dashboards, and spreadsheets. Both text and visual output (for example, charts, tables) can be displayed immediately or reported on operations. - Genbi Insights:
Wrenai provides AI-generated summary, reporting and context-aware visualization, allowing for fast, decision-ready analysis. - LLM Flexibility:
Wrenai supports a range of large language models, including:- Openai GPT series
- Azure Openai
- Google Gemini, Vertex AI
- DeepSeek
- Data screen
- AWS BedRock (anthropomorphic Claude, Cohere, etc.)
- valley
- Ollama (for deployment of local or custom LLM)
- Other OpenAI API compatible and user-defined models.
- Semantic layer and index:
Use the Modeling Definition Language (MDL) to encode patterns, metrics, joins, and definitions to provide accurate context for LLMS and reduce hallucinations. The semantics engine ensures context-rich queries, architectural embeddings, and correlation-based retrieval for accurate SQL. - Export and collaboration:
Results can be exported to Excel, Google Sheets, or API for further analysis or team sharing. - API embedding:
Query and visualization capabilities are accessible via the API, allowing seamless embedding in custom applications and front-ends.
Architecture Overview
Wrenai’s architecture is modular and can be used for powerful deployment and integration:
| Element | describe |
|---|---|
| user interface | A web-based or CLI UI for natural language query and data visualization. |
| Orchestra layer | Process input parsing, manage LLM selection and coordinate investigation and query execution. |
| Semantic Index | Embed database schema and metadata to provide critical context for LLM. |
| LLM Abstraction | The unified API integrates multiple LLM providers on the cloud and on-premises. |
| Query Engine | Execute generated SQL on supported databases/data warehouses. |
| Visualization | Render tables, charts, dashboards and export results. |
| Plug-in/Extensibility | Allows the integration of custom connectors, templates, prompt logic and domain-specific requirements. |
Semantic Engine Details
- Architecture embedding:
Dense vectors represent capture patterns and business environments, thus powering relevance-based retrieval. - Rarely launch facilitation and metadata injection:
Architecture samples, joining and business logic are injected into LLM prompts for improved reasoning and accuracy. - Context compression:
The engine adjusts the size of the pattern representation based on the token limit and retains key details for each model. - The Hound Generation:
Related schemas and metadata are collected through vector searches and added to context-aligned prompts. - Model Agility:
WREN Engine works on LLMS through protocol-based abstraction, ensuring a consistent context regardless of the backend.
Supported integrations
- Databases and repositories:
Out-of-the-box support for BigQuery, PostgreSQL, MySQL, Microsoft SQL Server, Clickhouse, Trino, Snowflake, DuckDB, Amazon Athena, and Amazon Redshift. - Deployment Mode:
Self-hosting can be done in the cloud or as a hosting service. - API and embedding:
Easily integrate into other applications and platforms through APIs.
Typical use cases
- Marketing/Sales:
Quickly generate performance charts, funnel analysis or region-based summary from natural language cues. - Products/Operations:
Analyze product usage, churn or operational metrics with follow-up questions and visual summary. - Executive/Analyst:
Automated, latest business dashboards and KPI tracking are delivered in minutes.
in conclusion
Wrenai is a proven open source Genbi solution that bridges the gap between business teams and databases through dialogue, context-aware, AI-driven analytics. It is scalable, multi-LLM compatible, secure and designed with a powerful semantic main chain to ensure trustworthy, interpretable and easy to integrate business intelligence.
Check Github page. All credits for this study are to the researchers on the project.
Join NVIDIA, OpenAI, DeepMind, Meta, Microsoft, JP Morgan Chase, Amgan, Amgan, Aflac, Aflac, Wells Fargo and 100s NVIDIA, OpenAI, Meta, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, Microsoft, and 100s researchers.
Asif Razzaq is CEO of Marktechpost Media Inc. As a visionary entrepreneur and engineer, ASIF is committed to harnessing the potential of artificial intelligence to achieve social benefits. His recent effort is to launch Marktechpost, an artificial intelligence media platform that has an in-depth coverage of machine learning and deep learning news that can sound both technically, both through technical voices and be understood by a wide audience. The platform has over 2 million views per month, demonstrating its popularity among its audience.
