What are the hardware requirements for Mpt 7B Q8?

Mpt 7B Q8 requires approximately 6.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mpt 7B Q8 and how large is it?

Mpt 7B Q8 is developed by Abzu, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Mpt 7B Q8 perform on standard benchmarks?

Mpt 7B Q8 has the following published scores: MMLU 42.58. Compare against reference models on this page or on the LLM Explorer leaderboards.

Mpt 7B Q8 by Abzu — VRAM 6.9GB

Name: Mpt 7B Q8
Author: Abzu

Mpt 7B Q8 is an open-source language model by Abzu. Features: 7b LLM, VRAM: 6.9GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.11, Arc: 47.4, HellaSwag: 77.4, MMLU: 42.6, GSM8K: 8.3.

Model Card on HF 🤗: https://huggingface.co/Abzu/mpt-7b-q8

Mpt 7B Q8 Parameters and Internals

Model Type

decoder-style transformer

Additional Notes

Utilizes ALiBi to handle extremely long inputs beyond its training context length. The model can be utilized with HuggingFace's Transformers library.

Supported Languages

English (Unknown)

Training Details

Data Sources:

mc4, c4, togethercomputer/RedPajama-Data-1T, bigcode/the-stack, allenai/s2orc

Data Volume:

1T tokens

Methodology:

Transformer architecture modifications including ALiBi and FlashAttention

Context Length:

2048

Training Time:

9.5 days

Hardware Used:

440 A100-40GB GPUs

Model Architecture:

Decoder-only transformer using FlashAttention and ALiBi

Input Output

Input Format:

Text

Accepted Modalities:

Text

Output Format:

Text

LLM Name	Mpt 7B Q8
Repository 🤗	https://huggingface.co/Abzu/mpt-7b-q8
Base Model(s)	Mpt 7B 8K mosaicml/mpt-7b-8k
Model Size	7b
Required VRAM	6.9 GB
Updated	2026-08-25
Maintainer	Abzu
Model Type	mpt
Model Files	6.9 GB
Quantization Type	q8
Model Architecture	MPTForCausalLM
License	apache-2.0
Model Max Length	2048
Transformers Version	4.30.2
Tokenizer Class	GPTNeoXTokenizer
Vocabulary Size	50432
Torch Data Type	bfloat16

Rank the Mpt 7B Q8 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 Mpt 7B Q8

Best Alternatives	Context / RAM	Downloads	Likes
Mpt 7B Chat Q8	0K / 6.9 GB	17	1
Mpt 7B Storywriter Q8	0K / 6.9 GB	14	2
Mpt 7B Instruct Q8	0K / 6.9 GB	18	2
Mpt 7B Storywriter 4bit 128g	0K / 3.9 GB	194	127
...writer 4bit 128g 65kTokens CPU	0K / 3.9 GB	15	9
Mpt 7B Storywriter 4bit 128g	0K / 3.9 GB	23	20
Mpt 7B Chat	0K / 13.3 GB	80920	518
Mpt 7B Assistant	0K / 13.3 GB	11	1
Mpt 7B	0K / 13.3 GB	18460	1173
Mpt 7B Instruct	0K / 13.3 GB	7946	470

Note: green Score (e.g. "73.2") means that the model is better than Abzu/mpt-7b-q8.