Instructions to use kulwamalyango98/kiki-tafsiri-b2-4b-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use kulwamalyango98/kiki-tafsiri-b2-4b-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/gemma-4-e4b-unsloth-bnb-4bit") model = PeftModel.from_pretrained(base_model, "kulwamalyango98/kiki-tafsiri-b2-4b-lora") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Desktop
kiki-b2-4b-lora
Model Details
Model Description
Kiki B2 4B LoRA is a Kiswahili-English translation adapter built on Gemma 4 E4B. It is designed for translation between Kiswahili and English, with support for terms, short phrases, sentences, and some longer formal text.
- Developed by: Kulwa Malyango
- Shared by: kulwamalyango98
- Model type: LoRA adapter for a causal language model
- Base model: google/gemma-4-E4B / unsloth/gemma-4-E4B
- Language(s): Kiswahili, English
- Library: PEFT / Transformers / Unsloth
- License: Gemma license
Uses
Direct Use
This adapter can be used for:
- Kiswahili to English translation
- English to Kiswahili translation
- Translation of short terms, phrases, sentences, and some formal text
Recommended prompt format:
Translate from Swahili to English. Output only the translation.
Swahili: <text>
English:
- Downloads last month
- 5
Model tree for kulwamalyango98/kiki-tafsiri-b2-4b-lora
Base model
google/gemma-4-E4B