Our Contribution
to Open-Source AI
Privacy-focused, fine-tuned open-source models that convert plain English into precise SQL. Run locally with Ollama or integrate via HuggingFace - no cloud required.
HuggingFace downloads
Hrida T2SQL 3B V0.1
Ollama downloads
hrida-t2sql · hrida-t2sql-128k
3B
Parameters
128k
Context
Open
License
Hrida-T2SQL
3B · Open Source
Show top customers by revenue...
SELECT c.name,
SUM(o.amount) AS rev
FROM customers c
ORDER BY rev DESC
HuggingFace
Ollama
ollama run
hrida-t2sql
Self-hosted
Our Models
A smarter way
to query data.
Three open-source models, each fine-tuned for progressively complex SQL generation from natural language.
“Show top 10 customers by revenue in Q4 2024”
SELECT c.name, SUM(o.amount) AS revenue FROM customers c JOIN orders o ON c.id = o.customer_id WHERE YEAR(o.created_at) = 2024 AND QUARTER(o.created_at) = 4 ORDER BY revenue DESC LIMIT 10
Capabilities
Built for real deployments.
Fine-tuned for accuracy. Designed to run anywhere - from your laptop to your production infra.
Privacy-first
Run entirely on-premise. Your data never leaves your environment.
Self-hosted
Deploy via Ollama on any machine in minutes. No cloud account needed.
3B parameters
Compact enough for consumer hardware, powerful enough for production SQL.
HuggingFace native
Integrate with the transformers library in a single pipeline call.
Quick Start
Running in
under a minute.
Two ways to get started - pull via Ollama for instant local use, or integrate with the HuggingFace transformers library for production.
3
Models published
3B
Parameters each
128k
Max context
Free
Open weights
via Ollama
Local deployment
via HuggingFace