What are the hardware requirements for Bloom 1b7 Gptq 4bit?

Bloom 1b7 Gptq 4bit requires approximately 2.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Bloom 1b7 Gptq 4bit and how large is it?

Bloom 1b7 Gptq 4bit is developed by iproskurina, a model with 1.2b 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 Bloom 1b7 Gptq 4bit?

Bloom 1b7 Gptq 4bit is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Bloom 1b7 Gptq 4bit by iproskurina — VRAM 2.7GB

Name: Bloom 1b7 Gptq 4bit
Author: iproskurina

Bloom 1b7 Gptq 4bit is an open-source language model by iproskurina. Features: 1.2b LLM, VRAM: 2.7GB, License: bigscience-bloom-rail-1.0, Quantized, LLM Explorer Score: 0.1.

Arxiv:2210.17323 4bit Ak Ar As Autotrain compatible Base model:bigscience/bloom-1b... Base model:finetune:bigscience... Bloom Bm Bn Ca Code Dataset:c4 En Es Eu Fon Fr Gptq Gu Hi Id Ig Ki Kn Lg Ln Ml Mr Ne Nso Ny Or Pa Pretrained Pt Quantized Region:us Rn Rw Safetensors Sn St Sw Ta Te Tn Ts Tum Tw Ur Vi Wo Xh Yo Zh Zhs Zht Zu

Model Card on HF 🤗: https://huggingface.co/iproskurina/bloom-1b7-gptq-4bit

Bloom 1b7 Gptq 4bit Parameters and Internals

Model Type

bloom

Additional Notes

The model has been quantized to 4bit using AutoGPTQ with all parameters taken from the GPTQ paper.

Supported Languages

ak (unknown), ar (unknown), as (unknown), bm (unknown), bn (unknown), ca (unknown), code (unknown), en (unknown), es (unknown), eu (unknown), fon (unknown), fr (unknown), gu (unknown), hi (unknown), id (unknown), ig (unknown), ki (unknown), kn (unknown), lg (unknown), ln (unknown), ml (unknown), mr (unknown), ne (unknown), nso (unknown), ny (unknown), or (unknown), pa (unknown), pt (unknown), rn (unknown), rw (unknown), sn (unknown), st (unknown), sw (unknown), ta (unknown), te (unknown), tn (unknown), ts (unknown), tum (unknown), tw (unknown), ur (unknown), vi (unknown), wo (unknown), xh (unknown), yo (unknown), zh (unknown), zhs (unknown), zht (unknown), zu (unknown)

Training Details

Data Sources:

C4 dataset

Methodology:

Quantization using GPTQ with 128 random 2048 token segments

Context Length:

2048

Input Output

Accepted Modalities:

text

Output Format:

text

LLM Name	Bloom 1b7 Gptq 4bit
Repository 🤗	https://huggingface.co/iproskurina/bloom-1b7-gptq-4bit
Model Name	bloom-1b7
Model Creator	bigscience
Base Model(s)	Bloom 1b7 bigscience/bloom-1b7
Model Size	1.2b
Required VRAM	2.7 GB
Updated	2024-09-23
Maintainer	iproskurina
Model Type	bloom
Model Files	2.7 GB
Supported Languages	ak ar as bm bn ca code en es eu fr gu hi id ig ki kn lg ln ml mr ne ny or pa pt rn rw sn st sw ta te tn ts tw ur vi wo xh yo zh zu
GPTQ Quantization	Yes
Quantization Type	gptq\|4bit
Model Architecture	BloomForCausalLM
License	bigscience-bloom-rail-1.0
Transformers Version	4.35.2
Tokenizer Class	BloomTokenizer
Padding Token	<pad>
Vocabulary Size	250880
Torch Data Type	float16

Rank the Bloom 1b7 Gptq 4bit 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 Bloom 1b7 Gptq 4bit

Best Alternatives	Context / RAM	Downloads	Likes
Bloom 1b7 GPTQ 4bit G128	0K / 2.7 GB	12	0

Note: green Score (e.g. "73.2") means that the model is better than iproskurina/bloom-1b7-gptq-4bit.