🌍Pan-African

Ultraner MCP

MCP

The official Ultraner MCP server lets AI agents create payments, refunds and checkout sessions and manage escrow through the Ultraner API across Africa. Built for agent-driven commerce workflows over the Ultraner payments platform.

WariMCP

MCP

An MCP server giving AI agents a unified interface to West African payment rails, wrapping CinetPay, Wave, Hub2/Ecobank, PAPSS, FedaPay, KKiaPay, Moneroo, MTN MoMo and Flutterwave. Runs as stdio for Claude/Cursor or as an Express HTTP API and exposes 8 tools including initiate_payment, verify_payment and list_providers.

Yellow Card

APIs

Licensed stablecoin payments infrastructure across 20+ African countries, offering a single REST API for cross-border payments, local fiat on/off-ramps, USD stablecoin savings and a Treasury Portal. SDKs in cURL/Python/Node/Ruby; converts USDT/USDC against NGN, KES, USD, EUR and 50+ currencies; KYB/KYC and sandbox required.

Zindi

AI

Africa's largest data science and AI competition platform where organizations host real-world ML challenges and a community of builders competes to solve them. Offers competitions, learning courses, jobs and leaderboards, with partners including Microsoft, Google, AWS and Google DeepMind.

afro.tools MCP

MCP

A remote HTTP MCP server (endpoint https://mcp.afro.tools) and Claude Code plugin that exposes a machine-readable registry of African API specs across 12+ providers (Flutterwave, LengoPay, Djomy and others) spanning payments, SMS, identity and logistics in 21 countries. Maintained by the afrotools org as an open community project.

stopwords-sw

Tools

A comprehensive Swahili (sw) stopwords collection distributed in JSON and text formats (npm/bower stopwords-sw) for text preprocessing in NLP pipelines. Maintained by the stopwords-iso project.

uroman

Tools

uroman is a universal romanizer that converts text in virtually any script to the Latin alphabet, with dedicated handling for Amharic and the Ge'ez/Ethiopic script. It also adds initial support for Coptic and processes script-native numerals.

Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning

Research

This paper introduces multilingual adaptive fine-tuning (MAFT) applied to 17 of the most-resourced African languages, producing the AfroXLMR family of models. Removing non-African-script tokens cuts model size by roughly 50 percent while matching the accuracy of single-language adaptation on named entity recognition, topic classification and sentiment analysis.

AfTerFibre African Fibre Optic Network Map

Datasets

Open geodata mapping terrestrial and undersea fibre optic cable routes across Africa, digitised from operator maps into GIS format. The dataset is downloadable as CSV, SHP, KML, GeoJSON and SVG via the map portal and GitHub, and is also mirrored in the FAO data catalog. It is community-maintained by Steve Song under an open-data model.

AfriCOMET

AI

AfriCOMET is a COMET-based machine translation evaluation model for African languages, scoring translation quality from source, hypothesis and reference triplets. The STL-1.1 version uses the afro-xlmr-large-76L encoder and was validated in the WMT 2024 Metrics Shared Task across 13 African-centric language pairs. It is released by the Masakhane community with reference-based and quality-estimation variants.

AfriHate Hate Speech Datasets

Datasets

Multilingual collection of hate speech and abusive language datasets covering 15 African languages, built from tweets annotated by native speakers. Each instance carries labels from 3 to 4 annotators with anonymous annotator IDs, downloadable on HuggingFace. Published at NAACL 2025.

AfriHate: A Multilingual Collection of Hate Speech and Abusive Language Datasets for African Languages

Research

AfriHate is a multilingual benchmark of hate speech and abusive language datasets covering 15 African languages, annotated by native speakers. The paper contributes classification baselines and hate speech and offensive language lexicons, and analyses why keyword-based moderation fails for low-resource African languages. It was released on arXiv in January 2025.

AfriHuBERT

AI

AfriHuBERT is a compact self-supervised speech representation model based on mHuBERT-147, continually pretrained via multilingual adaptive finetuning on over 10,000 hours of speech spanning more than 1,200 African languages and varieties. It improves spoken language identification and ASR over its base model and acts as an encoder for downstream African speech tasks. Its training data was aggregated from sources including BibleTTS, Kallaama, NaijaVoices and NCHLT.

AfriMMLU (IrokoBench)

Datasets

Human-translated multiple-choice question-answering evaluation benchmark covering 16 to 17 African languages plus English and French, derived from a subset of MMLU across subjects like maths, geography and law. Distributed as CSV and parquet on HuggingFace and forms part of the IrokoBench suite (MMLU, MGSM, XNLI). Licensed Apache 2.0.

AfriMT5

AI

AfriMT5 (afri-mt5-base) is an mT5-based machine translation model from the Masakhane community fine-tuned to translate across 16 African languages using multilingual adaptive fine-tuning. It targets news-domain translation for low-resource African languages, several of which were not previously covered by existing benchmarks. It is distributed as open weights on HuggingFace.

AfriMTE and AfriCOMET: Enhancing COMET to Embrace Under-resourced African Languages

Research

This paper introduces AfriMTE, human translation-evaluation data with simplified annotation guidelines for 13 African languages, and AfriCOMET, a learned machine translation quality metric built on an African-centric multilingual encoder. It addresses the difficulty of measuring translation progress for under-resourced African languages and reports improved correlation with human judgment over existing metrics.

AfriQA: Cross-lingual Open-Retrieval Question Answering for African Languages

Research

AfriQA is the first cross-lingual open-retrieval question answering benchmark for African languages, with more than 12,000 XOR-QA examples across 10 African languages. The paper shows that current automatic translation and multilingual retrieval methods perform poorly for these languages, where in-language digital content is scarce.

AfriSenti: A Twitter Sentiment Analysis Benchmark for African Languages

Research

AfriSenti is a sentiment analysis benchmark of more than 110,000 tweets in 14 African languages spanning four language families, annotated by native speakers. It underpinned SemEval-2023 Task 12, a shared task that attracted more than 200 participants, and documents data collection, annotation and baseline methods for low-resource languages.

Africa Electricity Transmission & Distribution Grid Map

Datasets

Open GIS map of existing and planned electricity transmission and distribution lines for all continental African countries and Madagascar, tagged by voltage (sub-kV to 700kV) and status. Four downloadable resources cover combined, existing-only and planned-only lines as GeoJSON plus a Shapefile package. Published by the World Bank on ENERGYDATA.INFO under CC BY 4.0 and last updated September 2024.

Africapolis Urban Agglomerations (OECD/SWAC)

Datasets

Standardised geospatial database of urban agglomerations across all 54 African countries, built from censuses, electoral registers and satellite imagery. Each urban area is a polygon with population and built-up-area attributes, available in 2015 and 2020 versions. Data is downloadable as geospatial vector files from africapolis.org under CC BY 4.0.

Afrobarometer Survey Data

Datasets

Nationally representative public-attitude surveys on democracy, governance, the economy and society across roughly 39 African countries and nine survey rounds. Country-level and merged multi-country datasets are provided with codebooks in SPSS, Stata and CSV. Data is free to use; the most recent Round 9 covers 2021 to 2023.

DataFirst Open Data Portal (UCT)

Datasets

Research data service cataloguing African household, labour and census microdata, centred on South Africa but including other African countries, with full survey documentation and questionnaires. Microdata downloads (Stata, SPSS, CSV) require a free account and agreement to terms, while the catalog itself is openly browsable. Maintained by the University of Cape Town.

Digital Earth Africa Analysis-Ready EO Data

Datasets

Continental analysis-ready satellite data platform providing cropland extent, Water Observations from Space, coastline change, fractional cover and GeoMAD products for all of Africa. Data is served as Cloud-Optimized GeoTIFF via a STAC API (explorer.digitalearth.africa/stac), OGC Web Services (ows.digitalearth.africa), an AWS S3 open bucket and a JupyterHub Sandbox. Products are derived from Landsat and Sentinel and are continuously updated.

Digital Public Infrastructure and Development: A World Bank Group Approach

Research

This World Bank Group paper sets out a shared approach to digital public infrastructure, framed around interoperable digital identity, digital payments and data-exchange layers as the foundation for inclusive digital economies. It provides definitions, guiding principles and safeguards to help countries plan and govern DPI. Much of the World Bank's DPI work targets low- and middle-income countries, including across Africa.