What are the hardware requirements for Stablelm Zephyr 3B?

Stablelm Zephyr 3B requires approximately 5.6 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Stablelm Zephyr 3B and how large is it?

Stablelm Zephyr 3B is developed by stabilityai, a model with 3b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Stablelm Zephyr 3B perform on standard benchmarks?

Stablelm Zephyr 3B has the following published scores: MMLU 46.17. Compare against reference models on this page or on the LLM Explorer leaderboards.

Stablelm Zephyr 3B by stabilityai — VRAM 5.6GB, 4K context

Name: Stablelm Zephyr 3B
Author: stabilityai

Stablelm Zephyr 3B is an open-source language model by stabilityai. Features: 3b LLM, VRAM: 5.6GB, Context: 4K, License: other, Instruction-Based, LLM Explorer Score: 0.23, Arc: 46.1, HellaSwag: 74.2, MMLU: 46.2, GSM8K: 42.2.

Arxiv:2305.18290 Arxiv:2306.05685 Conversational Dataset:huggingfaceh4/ultracha... Dataset:huggingfaceh4/ultrafee... Dataset:intel/orca dpo pairs Dataset:meta-math/metamathqa Dataset:wizardlm/wizardlm evol... En Endpoints compatible Instruct Model-index Region:us Safetensors Stablelm

Model Card on HF 🤗: https://huggingface.co/stabilityai/stablelm-zephyr-3b

Stablelm Zephyr 3B Benchmarks

ARC: 46.08 vs 96.7 (so35)^-52.3%

HellaSwag: 74.16 vs 95.3 (gpt4)^-22.2%

MMLU: 46.17 vs 88.3 (so35)^-47.7%

GSM8K: 42.15 vs 96.4 (so35)^-56.3%

LLME Score: 0.22661

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Stablelm Zephyr 3B (stabilityai/stablelm-zephyr-3b)

🌟 Advertise your project 🚀

Stablelm Zephyr 3B Parameters and Internals

LLM Name	Stablelm Zephyr 3B
Repository 🤗	https://huggingface.co/stabilityai/stablelm-zephyr-3b
Model Size	3b
Required VRAM	5.6 GB
Updated	2026-05-27
Maintainer	stabilityai
Model Type	stablelm
Instruction-Based	Yes
Model Files	5.6 GB
Supported Languages	en
Model Architecture	StableLmForCausalLM
License	other
Context Length	4096
Model Max Length	4096
Transformers Version	4.38.0
Tokenizer Class	GPTNeoXTokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	50304
Torch Data Type	bfloat16

Quantized Models of the Stablelm Zephyr 3B

Model	Likes	Downloads	VRAM
Stablelm Zephyr 3B GGUF	101	2034	1 GB
Stablelm Zephyr 3B GPTQ	12	109	1 GB

Best Alternatives to Stablelm Zephyr 3B

Best Alternatives	Context / RAM	Downloads	Likes
...t 3B Mix Spider Bird 200 Steps	16K / 5.6 GB	9	0
... Instruct 3B Spider 3500 Steps	16K / 11.2 GB	5	0
ReMask 3B	4K / 11.2 GB	97	16
Stable Code Instruct 3B 4bit	16K / 1.8 GB	52	5
Stable Cypher Instruct 3B	16K / 1.7 GB	625	33
Stable Code Instruct 3B	16K / 5.6 GB	1959	186
Stable Cypher Instruct 3B	16K / 1.7 GB	263	26
...struct 3B Spider 1500 Steps Q4	16K / 1.7 GB	83	0
Stable Code Instruct 3B AWQ	16K / 1.8 GB	6	0

Rank the Stablelm Zephyr 3B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

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

What open-source LLMs or SLMs are you in search of? 53999 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Stablelm Zephyr 3B by stabilityai