evidoria

← Back to browse

Good practice Imported

N-ATLAS — Nigeria's Open-Source AI Model for Yoruba, Hausa, Igbo and Nigerian-Accented English

Nigeria · Abuja · See the Nigeria profile · See the Abuja profile

Evidence: Descriptive / self-reported Top 62% 53/100 · Ask Evidence Copilot about this practice

Nigeria's NITDA and NCAIR, with Awarri Technologies, launched the open-source N-ATLAS language model at the UN General Assembly in September 2025, covering Yoruba, Hausa, Igbo and Nigerian-accented English so government chatbots can work in citizens' own languages.

400 million+
Training tokens
N-ATLAS — Nigeria's Open-Source AI Model for Yoruba, Hausa, Igbo and Nigerian-Accented English

Details

Maturity
Pilot
Promoter
National Information Technology Development Agency (NITDA) / National Centre for Artificial Intelligence and Robotics (NCAIR)
Period
2025–present
Keywords
digital government, language technology, AI infrastructure

Context

Nigeria's NITDA and NCAIR, with Awarri Technologies, released N-ATLAS — an open-source LLM fine-tuned on 400M+ tokens covering Yoruba, Hausa, Igbo and Nigerian-accented English — at UNGA80 in September 2025.

Results

As a newly released open model rather than a live evaluated service, there is no published data yet on translation accuracy, agency adoption or citizen-facing outcomes.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Short (< 1 year)
Staffing & skills
NITDA / NCAIR technical team, Awarri Technologies engineering team

Conditions for success

  • agency adoption in live citizen-facing chatbots/call centres
  • published accuracy evaluation post-deployment

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful