GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
uctnlp
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
uctnlp
Model Tree
Input Modalities
Output Modalities
Supported Functionality
125,008,384LlamaForCausalLM51215363093204810000.01e-5trueflash_attention_2MzansiLM uses a custom BPE tokenizer with a vocabulary size of 65536.
[BOS] = 0[EOS] = 1[PAD] = 2[UNK] = 3NFDByteLevel[BOS] $A [EOS][BOS] $A [EOS] [BOS] $B [EOS]The model was trained on MzansiText and covers all eleven official South African languages:
af, en, nso, sot, ssw, tsn, tso, ven, xho, zul, nbl
Related releases:
MzansiLM is a research model for pretraining, fine-tuning, and evaluation on South African languages. It is intended as a reproducible baseline for language modeling and downstream task adaptation.
MzansiLM uses an explicit attention head dimension (head_dim=56) with
hidden_size=512 and num_attention_heads=9. Use Transformers 4.x for this
release:
bash
pip install "transformers>=4.52.4,<5" torch accelerate safetensors
python
from transformers import AutoModelForCausalLM, AutoTokenizermodel_id = "anrilombard/mzansilm-125m"tokenizer = AutoTokenizer.from_pretrained(model_id)model = AutoModelForCausalLM.from_pretrained(model_id)inputs = tokenizer("Namuhla sifunda ngolimi ngoba abantu", return_tensors="pt")outputs = model.generate(**inputs,max_new_tokens=50,pad_token_id=tokenizer.pad_token_id,)print(tokenizer.decode(outputs[0], skip_special_tokens=True))
This is a base language model, not an instruction-tuned chat model, so outputs are plain text continuations of the prompt.
Transformers 5 currently rejects this Llama configuration during validation
because 512 is not divisible by 9, even though the model uses the explicit
head_dim=56 stored in config.json. Until that validation path supports this
configuration, use Transformers 4.x as shown above.
Please cite the paper:
bibtex
@misc{lombard2026mzansitextmzansilmopencorpus,title={MzansiText and MzansiLM: An Open Corpus and Decoder-Only Language Model for South African Languages},author={Anri Lombard and Simbarashe Mawere and Temi Aina and Ethan Wolff and Sbonelo Gumede and Elan Novick and Francois Meyer and Jan Buys},year={2026},eprint={2603.20732},archivePrefix={arXiv},primaryClass={cs.CL},url={https://arxiv.org/abs/2603.20732},}
Apache License 2.0