What are the hardware requirements for Opt 6.7B GPTQ 4bit G128?

Opt 6.7B GPTQ 4bit G128 requires approximately 4.2 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Opt 6.7B GPTQ 4bit G128 and how large is it?

Opt 6.7B GPTQ 4bit G128 is developed by iproskurina, a model with 6.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Opt 6.7B GPTQ 4bit G128 perform on standard benchmarks?

Opt 6.7B GPTQ 4bit G128 has the following published scores: MMLU 24.57. Compare against reference models on this page or on the LLM Explorer leaderboards.

Opt 6.7B GPTQ 4bit G128 by iproskurina — VRAM 4.2GB, 2K context

Name: Opt 6.7B GPTQ 4bit G128
Author: iproskurina

Opt 6.7B GPTQ 4bit G128 is an open-source language model by iproskurina. Features: 6.7b LLM, VRAM: 4.2GB, Context: 2K, License: other, Quantized, LLM Explorer Score: 0.14, Arc: 39.2, HellaSwag: 68.7, MMLU: 24.6, GSM8K: 1.

Arxiv:2210.17323 4-bit 4bit Base model:facebook/opt-6.7b Base model:quantized:facebook/... En Gptq Opt Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/iproskurina/opt-6.7b-GPTQ-4bit-g128

Opt 6.7B GPTQ 4bit G128 Parameters and Internals

Model Type

opt, text-generation

Additional Notes

Quantized by iproskurina using AutoGPTQ.

Supported Languages

en (High)

LLM Name	Opt 6.7B GPTQ 4bit G128
Repository 🤗	https://huggingface.co/iproskurina/opt-6.7b-GPTQ-4bit-g128
Model Name	opt-6.7b
Model Creator	facebook
Base Model(s)	facebook/opt-6.7b facebook/opt-6.7b
Model Size	6.7b
Required VRAM	4.2 GB
Updated	2026-07-19
Maintainer	iproskurina
Model Type	opt
Model Files	4.2 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq\|4bit
Model Architecture	OPTForCausalLM
License	other
Context Length	2048
Model Max Length	2048
Transformers Version	4.44.2
Tokenizer Class	GPT2Tokenizer
Padding Token	<pad>
Vocabulary Size	50272
Torch Data Type	float16
Activation Function	relu
Errors	replace

Rank the Opt 6.7B GPTQ 4bit G128 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 Opt 6.7B GPTQ 4bit G128

Best Alternatives	Context / RAM	Downloads	Likes
Opt 6.7B Gptq 4bit	2K / 4.2 GB	90	0
...7B Gqa Ub 16 Best For KV Cache	2K / 27.4 GB	190	0
...B Qcqa Ub 16 Best For KV Cache	2K / 27.4 GB	94	0
....7B Qcqa Ub 16 Best For Q Loss	2K / 27.4 GB	60	0
Galactica 6.7B ReFT GSM8k	2K / 13.8 GB	58	0
Galactica 6.7B EssayWriter	2K / 13.4 GB	55	5
Opt Flan Iml 6.7B	2K / 26.6 GB	131	1
Opt 6.7B	2K / 13.4 GB	120463	121
...lactica 6.7B Evol Instruct 70K	2K / 27.4 GB	129	19
OPT 6.7B Nerybus Mix	2K / 13.7 GB	125	20

Note: green Score (e.g. "73.2") means that the model is better than iproskurina/opt-6.7b-GPTQ-4bit-g128.