What are the hardware requirements for Lunar 4B?

Lunar 4B requires approximately 7.7 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Lunar 4B and how large is it?

Lunar 4B is developed by Sakalti, 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 Lunar 4B?

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

Lunar 4B by Sakalti — VRAM 7.7GB, 128K context

Name: Lunar 4B
Author: Sakalti

Lunar 4B is an open-source language model by Sakalti. Features: 4b LLM, VRAM: 7.7GB, Context: 128K, License: mit, Instruction-Based, Merged, LLM Explorer Score: 0.16.

Merged Model Arxiv:2306.01708 Base model:axcxept/borea-phi-3... Base model:finetune:axcxept/bo... Conversational Custom code En Endpoints compatible Instruct Ja Phi3 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Sakalti/Lunar-4B

Lunar 4B Parameters and Internals

LLM Name	Lunar 4B
Repository 🤗	https://huggingface.co/Sakalti/Lunar-4B
Base Model(s)	microsoft/phi-3.5-mini-instruct AXCXEPT/Borea-Phi-3.5-mini-Instruct-Common microsoft/phi-3.5-mini-instruct AXCXEPT/Borea-Phi-3.5-mini-Instruct-Common
Merged Model	Yes
Model Size	4b
Required VRAM	7.7 GB
Updated	2026-07-26
Maintainer	Sakalti
Model Type	phi3
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-2 2.7 GB: 2-of-2
Supported Languages	en ja
Model Architecture	Phi3ForCausalLM
License	mit
Context Length	131072
Model Max Length	131072
Transformers Version	4.46.2
Tokenizer Class	LlamaTokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	32064
Torch Data Type	float16

Rank the Lunar 4B 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 Lunar 4B

Best Alternatives	Context / RAM	Downloads	Likes
Phi4 Mini Prose2Tags 4B	128K / 7.7 GB	20	0
Calme 2.1 Phi3.5 4B	128K / 7.7 GB	18	4
Vanta Research Apollo V1 4B	128K / 7.7 GB	7	0
AURORAV0.3 4B	128K / 7.7 GB	25	3
Phi3.5 Gutenberg 4B	128K / 7.7 GB	13	4
Calme 2.1 Phi3 4B	4K / 7.7 GB	11	1
Calme 2.3 Phi3 4B	4K / 7.7 GB	66	9
Calme 2.2 Phi3 4B	4K / 7.7 GB	7	3
Ruphi 4B	128K / 7.6 GB	6	0
Phi 3.5 Mini Hyper	128K / 7.7 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than Sakalti/Lunar-4B.