What are the hardware requirements for Mai3?

Mai3 requires approximately 14.5 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mai3 and how large is it?

Mai3 is developed by mishikaai, 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 Mai3?

Mai3 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Mai3 by mishikaai — VRAM 14.5GB, 32K context

Name: Mai3
Author: mishikaai

Mai3 is an open-source language model by mishikaai. Features: 7b LLM, VRAM: 14.5GB, Context: 32K, Merged, LLM Explorer Score: 0.17.

Merged Model Base model:finetune:wizardlmte... Base model:wizardlmteam/wizard... Endpoints compatible Mistral Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mishikaai/Mai3

Mai3 Parameters and Internals

LLM Name	Mai3
Repository 🤗	https://huggingface.co/mishikaai/Mai3
Base Model(s)	Teknium/OpenHermes-2.5-Mistral-7B WizardLM/WizardMath-7B-V1.1 Teknium/OpenHermes-2.5-Mistral-7B WizardLM/WizardMath-7B-V1.1
Merged Model	Yes
Model Size	7b
Required VRAM	14.5 GB
Updated	2026-08-03
Maintainer	mishikaai
Model Type	mistral
Model Files	4.9 GB: 1-of-3 4.9 GB: 2-of-3 4.7 GB: 3-of-3
Model Architecture	MistralForCausalLM
Context Length	32768
Model Max Length	32768
Transformers Version	4.49.0
Tokenizer Class	LlamaTokenizer
Padding Token	<s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Mai3 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 Mai3

Best Alternatives	Context / RAM	Downloads	Likes
...Nemo Instruct 2407 Abliterated	1000K / 24.5 GB	220	20
MegaBeam Mistral 7B 512K	512K / 14.4 GB	8470	54
SpydazWeb AI HumanAI RP	512K / 14.4 GB	16	1
SpydazWeb AI HumanAI 002	512K / 14.4 GB	18	1
...daz Web AI ChatML 512K Project	512K / 14.5 GB	12	0
MegaBeam Mistral 7B 300K	282K / 14.4 GB	3779	16
MegaBeam Mistral 7B 300K	282K / 14.4 GB	8018	17
Hebrew Mistral 7B 200K	256K / 30 GB	23	15
Astral 256K 7B	250K / 14.4 GB	9	0
Astral 256K 7B V2	250K / 14.4 GB	8	0

Note: green Score (e.g. "73.2") means that the model is better than mishikaai/Mai3.