31+ Open-Weight Models

    Models built for real-world language

    A dedicated African-language NLP stack, tokenizers, encoders, embeddings, language identification, and diacritization, alongside frontier-merged LLMs, rerankers, generators, and Prism vision models. All open-weight, all on Hugging Face.

    Browse Families 🤗 View on Hugging Face
    quickstart.py
    from transformers import pipeline
    # Identify the language, then restore its tone marks
    lid = pipeline("text-classification", model="olaverse/lid-neural-25.2")
    print(lid("Kedu ka i mere?"))
    [{'label': 'ibo', 'score': 0.998}]

    31+

    Open Models

    9

    Model Families

    4,637+

    All-Time Downloads

    25+

    Languages Covered

    Browse by Family

    Every model, grouped by what it's built to do.

    5 models

    MIST

    Frontier-merged LLMs: from MIST-Mini-8B to MIST-1-140B, including reasoning-tuned Thinking variants.

    Explore
    6 models

    LID

    Language identification: from 5 Nigerian languages to 25 languages worldwide.

    Explore
    6 models

    DiacNet

    Diacritic & tone restoration for Yorùbá, Igbo, Hausa and beyond.

    Explore
    5 models

    Prism

    Vision models: super-resolution upscaling, denoising, and image steganography.

    Explore
    2 models

    MIST Embed

    Encoders & sentence embeddings: the retrieval foundation for Nigerian languages.

    Explore
    2 models

    MIST Rerank

    Cross-encoder rerankers, 22.7M to 150M: precision for search & RAG pipelines.

    Explore
    2 models

    MIST Gen

    Task generators: question generation in 25+ languages and title generation.

    Explore
    2 models

    OTK Tokenizers

    Byte-level BPE tokenizers optimised for Nigerian & African-language text.

    Explore
    1 model

    Legal Peace

    Fine-tuned legal reasoning and contract-analysis LLM.

    Explore

    All Models

    The complete index, live from Hugging Face, sorted by downloads.

    ModelFamilyTaskDownloadsLikesLink
    diacnet-1.0
    African-language diacritic and tone restoration powered by a multilingual ByT5 model, making text readable, searchable, and machine-usable.
    		DiacNetText-to-text1,306 0🤗 HF
    MIST-Mini-8B
    General-purpose 8B merged LLM built on the Llama 3.1 lineage with advanced model merging.
    		MISTText generation666 4🤗 HF
    MIST-1-70B
    General-purpose 70B merged LLM built on the Llama 3.1 lineage with advanced model merging.
    		MISTText generation513 1🤗 HF
    MIST-Mini-8B-Thinking
    Reasoning-tuned MIST model trained with GRPO to think step-by-step before answering.
    		MISTText generation291 3🤗 HF
    mist-tg-0.3b
    Title-generation model that names chats and documents from their content. Small, fast, and ByT5-based.
    		MIST GenText generation272 0🤗 HF
    legal-peace-v1.0
    Specialised legal reasoning model fine-tuned for contract analysis and document comprehension.
    		Legal PeaceText generation235 2🤗 HF
    MIST-1-140B
    General-purpose 140B frankenmerge LLM built on the Llama 3.1 lineage with advanced model merging.
    		MISTText generation183 0🤗 HF
    lid-neural-5
    Neural language identification across Nigerian languages (Yorùbá, Igbo, Hausa, Pidgin, English).
    		LIDClassification168 0🤗 HF
    MIST-1-140B-4bit
    4-bit NF4 quantised release of MIST-1-140B: frontier-scale weights on accessible hardware.
    		MISTText generation160 0🤗 HF
    naija-embed-base
    Cross-lingual sentence-embedding model for Hausa, Yorùbá, and Igbo. It is the retrieval foundation for Nigerian-language search and RAG.
    		MIST EmbedEmbeddings114 0🤗 HF
    lid-neural-5.1
    Neural language identification across Nigerian languages (Yorùbá, Igbo, Hausa, Pidgin, English).
    		LIDClassification105 0🤗 HF
    prism-steganography
    Neural steganography model that invisibly hides data inside images and recovers it losslessly.
    		PrismImage-to-image85 0🤗 HF
    mist-reranker-150m
    Cross-encoder reranker (150M-parameter ModernBERT) that re-orders retrieved passages by true relevance for search and RAG.
    		MIST RerankReranking77 0🤗 HF
    lid-lite-25
    fastText language identification across 25 world languages.
    		LIDClassification64 0🤗 HF
    lid-neural-25.1
    Neural language identification across 25 world languages.
    		LIDClassification61 0🤗 HF
    lid-neural-25.2
    Neural language identification across 25 world languages.
    		LIDClassification59 0🤗 HF
    mist-qg-1.5b
    Question-generation model that turns any passage into natural questions across 25+ languages.
    		MIST GenGeneration55 0🤗 HF
    mist-encoder-base-ng
    ModernBERT masked-LM encoder pretrained for Hausa, Yorùbá, Igbo, and Nigerian Pidgin. A foundation for Nigerian-language NLP.
    		MIST EmbedMasked LM53 0🤗 HF
    prism-upscaler-2x
    Super-resolution upscaler delivering 2x image enlargement with sharp, artefact-free detail.
    		PrismImage-to-image39 0🤗 HF
    mist-reranker-22.7M
    Cross-encoder reranker (22.7M-parameter micro) that re-orders retrieved passages by true relevance for search and RAG.
    		MIST RerankReranking38 0🤗 HF
    prism-upscaler-max
    Super-resolution upscaler delivering maximum-quality image enlargement with sharp, artefact-free detail.
    		PrismImage-to-image34 0🤗 HF
    prism-upscaler-4x
    Super-resolution upscaler delivering 4x image enlargement with sharp, artefact-free detail.
    		PrismImage-to-image33 0🤗 HF
    prism-denoiser
    Image denoising model that cleans noisy photos while preserving fine detail.
    		PrismImage-to-image26 0🤗 HF
    otk-bpe-50k
    Byte-level BPE tokenizer built for Nigerian languages, Yorùbá, Igbo, Hausa, and Pidgin, efficient subword segmentation where general vocabularies fall apart.
    		OTK TokenizersTokenizer0 0🤗 HF
    lid-lite-5
    Zero-dependency language identification across Nigerian languages (Yorùbá, Igbo, Hausa, Pidgin, English).
    		LIDLanguage ID0 0🤗 HF
    diacnet-yor-viterbi
    Yorùbá diacritic and tone restoration powered by Viterbi decoding, making text readable, searchable, and machine-usable.
    		DiacNetDiacritization0 0🤗 HF
    diacnet-yor-db
    Yorùbá diacritic and tone restoration powered by k-nearest neighbours, making text readable, searchable, and machine-usable.
    		DiacNetDiacritization0 0🤗 HF
    diacnet-ig
    Igbo diacritic and tone restoration powered by k-nearest neighbours, making text readable, searchable, and machine-usable.
    		DiacNetDiacritization0 0🤗 HF
    diacnet-yor
    Yorùbá diacritic and tone restoration powered by BiLSTM networks, making text readable, searchable, and machine-usable.
    		DiacNetDiacritization0 0🤗 HF
    diacnet-yor-x
    Yorùbá diacritic and tone restoration powered by transformer encoders, making text readable, searchable, and machine-usable.
    		DiacNetDiacritization0 0🤗 HF
    otk-bpe
    Byte-level BPE tokenizer built for multilingual text, efficient subword segmentation where general vocabularies fall apart.
    		OTK TokenizersTokenizer0 0🤗 HF

    Featured Models

    A cross-section of what's live today: one from each corner of the stack.

    MIST

    MIST-1-140B

    General-purpose 140B frankenmerge LLM built on the Llama 3.1 lineage with advanced model merging.

    183 0
    View
    DiacNet

    diacnet-1.0

    African-language diacritic and tone restoration powered by a multilingual ByT5 model, making text readable, searchable, and machine-usable.

    1,306 0
    View
    LID

    lid-neural-25.2

    Neural language identification across 25 world languages.

    59 0
    View
    Prism

    prism-upscaler-4x

    Super-resolution upscaler delivering 4x image enlargement with sharp, artefact-free detail.

    33 0
    View

    Build With Our Models

    Every model here is open-weight and free to use. Fine-tune it, deploy it, or drop it straight into your own product.

    View on Hugging FaceFollow on GitHub