What are the hardware requirements for PyLlama 7B?

PyLlama 7B requires approximately 13.5 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed PyLlama 7B and how large is it?

PyLlama 7B is developed by mlabonne, a model with 7b 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 PyLlama 7B?

PyLlama 7B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

PyLlama 7B by mlabonne — VRAM 13.5GB

Name: PyLlama 7B
Author: mlabonne

PyLlama 7B is an open-source language model by mlabonne. Features: 7b LLM, VRAM: 13.5GB, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.08.

Dataset:mlabonne/evol-instruct... Endpoints compatible Instruct Llama Lora Pytorch Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlabonne/PyLlama-7b

PyLlama 7B Parameters and Internals

Model Type

text generation

Additional Notes

The model is fine-tuned using QLoRA with 4-bit precision on the Evol-Instruct-Python-26k dataset. The entire training is done on an RTX 3090 card.

Training Details

Data Sources:

mlabonne/Evol-Instruct-Python-26k

Methodology:

Fine-tuning using QLoRA (4-bit precision)

Context Length:

2048

Training Time:

9h 52m 34s

Hardware Used:

RTX 3090

Model Architecture:

LlamaForCausalLM

LLM Name	PyLlama 7B
Repository 🤗	https://huggingface.co/mlabonne/PyLlama-7b
Model Size	7b
Required VRAM	13.5 GB
Updated	2026-08-08
Maintainer	mlabonne
Instruction-Based	Yes
Model Files	0.2 GB 10.0 GB: 1-of-2 3.5 GB: 2-of-2 10.0 GB: 1-of-2 3.5 GB: 2-of-2 0.0 GB
Model Architecture	AutoModelForCausalLM
License	apache-2.0
Is Biased	none
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	k_proj\|v_proj\|o_proj\|gate_proj\|down_proj\|up_proj\|q_proj
LoRA Alpha	16
LoRA Dropout	0.05
R Param	32

Rank the PyLlama 7B 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 PyLlama 7B

Best Alternatives	Context / RAM	Downloads	Likes
Mistral 7B Instruct V0.3	32K / 14.5 GB	91	6
Mistral 7B Instruct V0.3	32K / 14.5 GB	1122	3
...ralai Mistral 7B Instruct V0.3	32K / 14.5 GB	6	3
Mistral 7B Instruct V0.3	32K / 14.5 GB	6	0
Full V4 Astromistral Final	32K / 4.5 GB	6	1
Dekillnovelist V2	0K / 0.2 GB	203	0
Mistral 7B Manumit V1	0K / 14.5 GB	130	0
Mistral 7B Instruct Uz	0K / 14.5 GB	59	15
Qwen Medical 7B	0K / 15.4 GB	24	0
Acemath Convfinqa	0K / 15.2 GB	6	0

Note: green Score (e.g. "73.2") means that the model is better than mlabonne/PyLlama-7b.