What are the hardware requirements for Kw5 149M?

Kw5 149M requires approximately 0.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Kw5 149M and how large is it?

Kw5 149M is developed by regnant-io, a model with 149m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Kw5 149M?

Kw5 149M is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Kw5 149M by regnant-io — VRAM 0.7GB

Name: Kw5 149M
Author: regnant-io

Kw5 149M is an open-source language model by regnant-io. Features: 149m LLM, VRAM: 0.7GB, License: apache-2.0, LLM Explorer Score: 0.33.

Arxiv:2512.17351 African-languages Base-model Kiswahili Kw5v2 Low-resource Pretrained Region:us Safetensors Sw Swahili

Model Card on HF 🤗: https://huggingface.co/regnant-io/kw5-149M

Kw5 149M Parameters and Internals

LLM Name	Kw5 149M
Repository 🤗	https://huggingface.co/regnant-io/kw5-149M
Model Size	149m
Required VRAM	0.7 GB
Updated	2026-09-18
Maintainer	regnant-io
Model Type	kw5v2
Model Files	0.7 GB
Supported Languages	sw
Model Architecture	KW5V2ForCausalLM
License	apache-2.0
Tokenizer Class	SentencePieceProcessor
Padding Token	<pad>
Vocabulary Size	32000
Torch Data Type	float32

Rank the Kw5 149M capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation