MORENA 0.2B nano
A 209M-parameter model pruned from MORENA 1.5B base and distilled from the 1.5B teacher on 10.5B tokens (155 A100 GPU-hours), built not as an assistant but as a cheap trunk for ASR rescoring, on-device keyboards and text normalisation in twelve African languages. Not a chat model, has no instruct version, not safety-tested.
| MORENA 0.2B nano | reference | |
|---|---|---|
| African bits per byte, mean of 12 | 1.583 | mzansilm-125m 2.464, SmolLM2-135M 3.570 |
| best under 0.5B parameters on | the same 11 languages as the mini | exception isiNdebele: mzansilm-125m 2.783 against the nano's 2.950 |
| decode speed, one A100 | 104 tokens/s, 2.31 ms/char | 58x cheaper per byte than Lugha-Llama-8B, the best outside model on bpb |
| Translation chrF++, en to African / African to en, 3-shot | 16.2 / 21.0 |
A GGUF build is in vamboai/morena-0.2b-nano-gguf.
The MORENA family
| model | params | African bpb (all 12, lower is better) | role |
|---|---|---|---|
| MORENA 1.5B base | 1.485B | 1.408 | pretrained and mid-trained; fine-tuning starting point |
| MORENA 1.5B instruct | 1.485B | 1.441 | chat, translation, tool calling; the model described in the paper |
| MORENA 0.5B mini | 503M | 1.520 | pruned and distilled from the 1.5B base |
| MORENA 0.5B mini instruct | 503M | 1.540 | chat fine-tune of the mini |
| MORENA 0.2B nano | 209M | 1.583 | cheap trunk for ASR rescoring, keyboards, normalisation |
26 models in total were measured on African bits per byte, 21 of them outside models from 125M to 12B parameters. MORENA 1.5B base is lowest of all 26 at 1.408; the 8B African specialist Lugha-Llama-8B is second at 1.423, ahead of the other four MORENA sizes on the twelve-language mean. Every general-purpose model measured, from 1B to 12B parameters, trails well behind: the best of them, gemma-3-12b-it, reaches only 2.159. Twelve languages: Shona, Swahili, Hausa, Yoruba, Igbo, isiZulu, isiXhosa, Kinyarwanda, Setswana, Afrikaans, isiNdebele and Nigerian Pidgin, plus English, French and code. isiNdebele (ISO code nbl) has no FLORES+ or MasakhaNEWS coverage and is evaluated on NCHLT transcripts only.
Tokenizer: 65,536-entry byte-fallback BPE trained on the target mix. African text costs 0.249 tokens per byte against 0.234 for English, about 6% more per byte than English in MORENA's vocabulary, but that same African text needs 1.39x fewer tokens than under Gemma 3's vocabulary and 1.53x fewer than under Llama 3.2's.
Translation is measured for seventeen models in total, thirteen of them outside models.
Author and citation
Isheanesu Misi, Vambo AI. Trained on CINECA Leonardo, with support from the AI Hub for Sustainable Development.
@techreport{misi2026morena,
title = {MORENA: An African Foundation Model},
author = {Misi, Isheanesu},
institution = {Vambo AI},
year = {2026},
month = {September},
note = {Released 18 September 2026}
}
Licence
Apache 2.0. Commercial use is permitted, including fine-tuning and redistributing derivatives, under the terms of the licence.
The training corpus is not released with these weights.