What are the hardware requirements for Llama 13B 4bit Hf?

Llama 13B 4bit Hf requires approximately 7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Llama 13B 4bit Hf and how large is it?

Llama 13B 4bit Hf is developed by 4bit, a model with 13b 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 Llama 13B 4bit Hf?

Llama 13B 4bit Hf is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Llama 13B 4bit Hf by 4bit — VRAM 7GB

Name: Llama 13B 4bit Hf
Author: 4bit

Llama 13B 4bit Hf is an open-source language model by 4bit. Features: 13b LLM, VRAM: 7GB, License: other, Quantized, LLM Explorer Score: 0.06.

4bit Endpoints compatible Llama Pytorch Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/4bit/llama-13b-4bit-hf

Llama 13B 4bit Hf Parameters and Internals

Model Type

auto-regressive, transformer

Use Cases

Areas:

research

Applications:

question answering, natural language understanding, reading comprehension

Primary Use Cases:

Developing techniques to improve language models

Limitations:

Should not be used for downstream applications without risk evaluation

Considerations:

The model can generate toxic content, incorrect information, or unhelpful answers due to lack of human feedback training.

Additional Notes

Requires special support code and converted via GPTQ method, experimental release.

Supported Languages

en (English), bg (Bulgarian), ca (Catalan), cs (Czech), da (Danish), de (German), es (Spanish), fr (French), hr (Croatian), hu (Hungarian), it (Italian), nl (Dutch), pl (Polish), pt (Portuguese), ro (Romanian), ru (Russian), sl (Slovene), sr (Serbian), sv (Swedish), uk (Ukrainian)

Training Details

Data Sources:

CCNet, C4, GitHub, Wikipedia, Books, ArXiv, Stack Exchange

Model Architecture:

transformer

Safety Evaluation

Risk Categories:

misinformation, bias, toxicity

Ethical Considerations:

The model reflects biases from its training data, which is collected from various web sources.

Responsible Ai Considerations

Fairness:

Evaluated on RAI datasets for gender, religion, race, sexual orientation, age, nationality, disability, physical appearance, and socio-economic status biases.

Transparency:

Bias evaluation results are reported.

Mitigation Strategies:

Filtered web data based on its proximity to Wikipedia using Kneser-Ney and fastText linear classifier.

Input Output

Accepted Modalities:

text

LLM Name	Llama 13B 4bit Hf
Repository 🤗	https://huggingface.co/4bit/llama-13b-4bit-hf
Base Model(s)	ToddLora 13b V2 autobots/ToddLora_13b_v2
Model Size	13b
Required VRAM	7 GB
Updated	2026-08-25
Maintainer	4bit
Model Type	llama
Model Files	7.0 GB
Quantization Type	4bit
Model Architecture	LLaMAForCausalLM
License	other
Transformers Version	4.27.0.dev0
Vocabulary Size	32000
Torch Data Type	float16

Rank the Llama 13B 4bit Hf 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 Llama 13B 4bit Hf

Best Alternatives	Context / RAM	Downloads	Likes
... X Alpaca 13B Native 4bit 128g	0K / 7.9 GB	106	727
... X Alpaca 13B Native 4bit 128g	0K / 8.1 GB	10	2
Llama 13B 4bit Gr128	0K / 7.5 GB	11	2
... X Alpaca 13B Native 4bit 128g	0K / 7.9 GB	33	3
Llama 13B 3bit Gr128	0K / 5.9 GB	9	1
Llama 13B 4bit Gr128	0K / 7.5 GB	7	5
Llama 13B 3bit Gr128	0K / 5.9 GB	10	3
Llm Jp 13B V2.0	4K / 27.4 GB	16	15
Swallow 13B GPTQ	4K / 7.5 GB	7	0
Alpaca 13B	0K / 52.1 GB	73	108

Note: green Score (e.g. "73.2") means that the model is better than 4bit/llama-13b-4bit-hf.