What are the hardware requirements for Sailor 4B?

Sailor 4B requires approximately 7.9 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Sailor 4B and how large is it?

Sailor 4B is developed by sail, a model with 4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Sailor 4B perform on standard benchmarks?

Sailor 4B has the following published scores: MMLU 37.45. Compare against reference models on this page or on the LLM Explorer leaderboards.

Sailor 4B by sail — VRAM 7.9GB, 32K context

Name: Sailor 4B
Author: sail

Sailor 4B is an open-source language model by sail. Features: 4b LLM, VRAM: 7.9GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.15, Arc: 43.9, HellaSwag: 69.5, MMLU: 37.5, GSM8K: 8.8.

Arxiv:2404.03608 Base model:finetune:qwen/qwen1... Base model:qwen/qwen1.5-4b Conversational Dataset:allenai/madlad-400 Dataset:cc100 Dataset:cerebras/slimpajama-62... Dataset:skywork/skypile-150b En Id Lo Model-index Ms Multilingual Qwen2 Region:us Safetensors Sailor Sea Th Vi Zh

Model Card on HF 🤗: https://huggingface.co/sail/Sailor-4B

Sailor 4B Parameters and Internals

LLM Name	Sailor 4B
Repository 🤗	https://huggingface.co/sail/Sailor-4B
Base Model(s)	Qwen/Qwen1.5-4B Qwen/Qwen1.5-4B
Model Size	4b
Required VRAM	7.9 GB
Updated	2026-07-09
Maintainer	sail
Model Type	qwen2
Model Files	7.9 GB
Supported Languages	en zh id th vi ms lo
Model Architecture	Qwen2ForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.37.2
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the Sailor 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 Sailor 4B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen1.5 4B Chat	32K / 7.9 GB	13877	46
Qwarkstar 4B Instruct Preview	32K / 9 GB	2	2
Qwen1.5 4B	32K / 7.9 GB	16820	36
Qwarkstar 4B	32K / 9 GB	5	0
Qwarkstar 4B Instruct	32K / 9 GB	30	1
Sailor 4B Chat	32K / 7.9 GB	155	2
Nusantara 4B Indo Chat	32K / 7.9 GB	59	2
Reyna CoT 4B V0.1	32K / 7.9 GB	103	6
Blossom V5 4B	32K / 7.9 GB	55	2
Quyen V0.1	32K / 7.9 GB	229	12

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