/ Use cases

    What our production models actually do

    Worked pipelines, each built only on our production models: maintained, benchmarked, and the ones we'd point anyone to today. Real code and published numbers.

    $pip install olaverse[deeplearning]
    quickstart.py
    from olaverse import LIDNeural608
    from olaverse.nlp import Diacritizer
    
    lid = LIDNeural608()
    lid.predict("Habari za asubuhi")         # → 'swh_Latn'
    
    d = Diacritizer(model="diacnet-2.0", lang="yor")
    d.restore("O so fun ara re pe oun ko ni isoro kankan.")
    # → 'Ó sọ fún ara rẹ̀ pé òun kò ní ìṣòro kankan.'

    Every pipeline below runs on the same SDK — read the docs

    10
    WORKED PIPELINES
    608
    LANGUAGES IDENTIFIED
    11
    LANGUAGES DIACRITIZED
    £0
    LICENCE COST
    / Pick your problem

    Start from the job, not the model

    A customer support email about a double charge
    / Decisions · Support & operations

    Automated decisions

    Route, flag and score tickets, chats and records with calibrated probabilities, in English and eight African languages, from one self-hosted model instead of a chain of LLM calls.

    PurpleMIST-Flash-1.0PurpleMIST-Mini-1.0
    A freight invoice ready for extraction
    / Extraction · Documents

    Document extraction

    Fill a JSON template from a photo, a scan or plain text, with a calibrated confidence on every field. Built for African documents and currencies, in English and African languages.

    MIST-Extract-1.0MIST-Extract-Flash-1.0
    A banking app screenshot of a failed transfer
    / Extraction · Decisions

    Payment messages

    Pull the amount, sender, reference and date out of mobile-money and bank SMS, then decide in one pass whether it is genuine, which account it belongs to and whether anyone needs to look at it.

    MIST-Extract-1.0PurpleMIST-Flash-1.0
    Text typed without marks, restored with its diacritics highlighted
    / Text quality

    Diacritic restoration

    Restore diacritics and tone marks on text that was typed, scraped, or keyed in without them, across 11 languages from Yorùbá and Igbo to Vietnamese, Polish and Arabic.

    diacnet-2.0diacnet-mini-2.0diactag-2.0
    Short texts in many languages, each labelled with its language
    / Routing & moderation

    Language routing

    Detect the language of support tickets, chat messages, queries, and documents across 608 languages, built African-first and accurate on the short text real products see.

    lid-neural-608lid-lite-608
    A passage and the questions generated from it
    / Content generation

    Question generation

    Generate the questions a passage answers, in 324 languages, in its own language or across languages, for quizzes, reading comprehension, help-centre FAQs and assessment banks.

    MIST-QG-1.1
    An Arabic sentence before and after its vowels are restored
    / Text quality · Arabic

    Arabic diacritization

    Restore Arabic tashkīl, with or without case endings, for text-to-speech, teaching material and readable copy, from models that also cover ten other languages.

    diactag-2.0diacnet-2.0
    A search typed without marks matching properly written Yorùbá
    / Search & retrieval

    Search that matches

    Restore the marks your documents lost, then match on an accent-folded key, so a user typing "oko" finds "ọkọ" and still sees it written properly.

    diactag-2.0lid-lite-608
    Crawled lines labelled by language, with noise dropped
    / Data engineering

    Low-resource corpus mining

    Sort billions of web documents into 608 languages on CPUs, drop the junk, and pull out the African and low-resource text that general detectors misfile.

    lid-lite-608lid-neural-608
    Query and passage pairs generated for retrieval training
    / ML engineering

    Retrieval training data

    Generate verified (question, passage) pairs from your own corpus in 324 languages, or across languages, and fine-tune retrievers and rerankers on data that matches your domain.

    lid-lite-608MIST-QG-1.1

    Not seeing yours? These are the pipelines we can point at published benchmarks for. Anything adjacent — a different language, a different task, your own data — we build to order.

    Want one of these in your product?

    Every model here is open-weight, so you can build any of it yourself today. If you would rather we adapted it to your data and shipped it, that is what Olaverse for Business does.