Overview
About our text and instruction datasets
Instruction and conversation data written by native speakers and experts, ready for fine-tuning and evaluation.
Where it's used
- LLM fine-tuning
- Indic assistants
- Evaluation sets
- Prototyping
What we offer
Instruction pairs
Prompts with high-quality, reviewed answers.
Conversations
Multi-turn chats in Indian languages and English.
Preference pairs
Chosen and rejected responses with reasons.
JSONL delivery
One record per line, ready for training pipelines.
How it works
From first call to final delivery
Scope
We agree the data, the quality bar, the formats and a sample size, and write it into clear guidelines.
Pilot
A small batch first. You review it, we tune the guidelines, and only then do we scale.
Produce
Vetted contributors work in our studios. Automatic checks catch problems as files arrive.
Verify
Consensus between annotators, hidden check items and reviewer audits on every batch.
Deliver
Clean exports in your format, with documentation and a quality report.
What you receive
- Sample files before purchase
- A datasheet for every dataset
- Commercial licence options
- Delivery by secure download
Industries we serve
Off-the-shelf datasets
More off-the-shelf datasets services
Speech corporaRead and conversational speech with transcripts and speaker details.Learn more Image datasetsLabelled photos of products, scenes and documents, ready to train on.Learn more Video datasetsActivity, traffic and indoor video, with optional labels.Learn more Egocentric video datasetsFirst-person video of household and work tasks, with hands in view.Learn more Custom dataset extensionsMore hours, another language or extra labels on an existing dataset.Learn more