What are the hardware requirements for Lythri 4B A2B?

Lythri 4B A2B requires approximately 9.3 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Lythri 4B A2B and how large is it?

Lythri 4B A2B is developed by Lythri, a model with 4b 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 Lythri 4B A2B?

Lythri 4B A2B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Lythri 4B A2B by Lythri — VRAM 9.3GB

Name: Lythri 4B A2B
Author: Lythri

Lythri 4B A2B is an open-source language model by Lythri. Features: 4b LLM, VRAM: 9.3GB, License: apache-2.0, LLM Explorer Score: 0.38.

Base model:finetune:google/gem... Base model:google/gemma-4-e2b Companion Conversational Emotional-support Endpoints compatible Eval-results Gemma4 Image-text-to-text On-device Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/Lythri/Lythri-4B-A2B

Lythri 4B A2B Parameters and Internals

LLM Name	Lythri 4B A2B
Repository 🤗	https://huggingface.co/Lythri/Lythri-4B-A2B
Base Model(s)	Gemma 4 E2B google/gemma-4-E2B
Model Size	4b
Required VRAM	9.3 GB
Updated	2026-10-04
Maintainer	Lythri
Model Type	gemma4
Model Files	9.3 GB
Model Architecture	Gemma4ForConditionalGeneration
License	apache-2.0
Transformers Version	5.16.1
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>

Rank the Lythri 4B A2B 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

Best Alternatives to Lythri 4B A2B

Best Alternatives	Context / RAM	Downloads	Likes
...B Sft Quality 20260927 86 Best	0K / 15.9 GB	318	0
CloudSurf 4B FC	0K / 15.9 GB	1321	2
En Indic Translate 4B	0K / 15.9 GB	564	0
Darwin 4B Genesis	0K / 15 GB	672	29
Tema Q X4 4B Thinking	0K / 10.2 GB	15	1
GhostShell 4B	0K / 15.9 GB	86	2