EU AI ACTto the 2 December 2027 Annex III enforcement deadline.Check your tier →
Home · Models · Bhasha 1.0
Sovereign model family · Indic voice & vernacular

Offline voice Q&A in Indian languages on a $30-class device. No internet, no latency, no tokens.

Bhasha is the Indic voice family: speech-in, speech-out question answering in Indian languages, small enough to run fully offline on edge hardware in the $30 class. Built for schools, field services and citizen-facing programmes where connectivity, latency and data residency all argue against the cloud.

POC Phase
Status
v1.0
Current version, frozen per deployment
4-bit
Quantised for your hardware
0
Outbound API calls
POC Phase
Bh
MindMap Bhasha 1.0
Bhasha — simply, language.
Tuned for

Offline Indic voice Q&A · vernacular text · edge devices

Classroom & learning assistants

Spoken Q&A on curriculum content in the student's own language, on a device that never needs a network.

Field-service voice

Hands-free, offline guidance for technicians and health workers in vernacular languages.

Citizen-service kiosks

Scheme eligibility, forms and instructions answered by voice at the point of service.

Vernacular document Q&A

Regional-language notices, circulars and forms made conversational.

Runs where the data lives

Edge devices, fully offline — 4-bit quantised for $30-class hardware; also deployable on-premise.

In production

POC phase; under evaluation for a state-scale school deployment.

References are anonymised by default; named references on request under NDA.

Proof point

Bhasha 1.0 vs Google Gemma 4 — head-to-head on Indic voice tasks

Composite task accuracy on an Indic-language task suite: the questions and document types our clients actually pay for, not a generic leaderboard.

Internal evaluation, August 2026. Gemma 4 run with published weights, same prompts and hardware. Golden set per domain, blind human + AI-judge scoring. Methodology and raw runs available on request.

Bhasha 1.0
92.1
Gemma 4
Google open model
86.4
+6% ahead, same prompts, same hardware
Governance

What comes with every Bhasha deployment

Provenance register

Base model, version and licence recorded for every deployment — the first artefact an auditor or a procurement team asks for.

Golden set per domain

Every family carries a curated evaluation set drawn from the document types its clients run on. Releases are gated on it.

Blind human + AI-judge scoring

Outputs scored by domain reviewers without knowing which model produced them, cross-checked by an AI judge. Raw runs available on request.

Frozen in production

The model that served your customers on Friday is the model serving them on Monday. No silent upgrades, no vendor deprecations, no pricing changes.

Plainly: Bhasha is a domain fine-tune of a permissively-licensed open-weight base, not a from-scratch foundation model. The base, its version and its licence are recorded in the provenance register for every deployment.
Evaluate before you commit

Run Bhasha on your own documents.

Send 20–50 representative documents or transcripts — anonymised is fine — and the acceptance rules your team uses today. We run Bhasha 1.0 on infrastructure you control, score it against your criteria, and hand back the results and the harness. Two weeks, fixed scope, no production data.

An engineer replies within one business day. Evaluations run on infrastructure you control; we never ask for production data.

Talk to the product team