What are the hardware requirements for Toffee 3B Beta?

Toffee 3B Beta requires approximately 7.7 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Toffee 3B Beta and how large is it?

Toffee 3B Beta is developed by maxzt, a model with 3b 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 Toffee 3B Beta?

Toffee 3B Beta is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Toffee 3B Beta by maxzt — VRAM 7.7GB, 8K context

Name: Toffee 3B Beta
Author: maxzt

Toffee 3B Beta is an open-source language model by maxzt. Features: 3b LLM, VRAM: 7.7GB, Context: 8K, License: mit, Instruction-Based, LLM Explorer Score: 0.31.

3b Base model:finetune:mistralai/... Base model:mistralai/ministral... Chat Conversational Dataset:huggingfaceh4/ultracha... Dataset:m-a-p/codefeedback-fil... Dataset:nvidia/openmathinstruc... En Instruct Ministral Mistral Model-index Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/maxzt/toffee-3b-beta

Toffee 3B Beta Parameters and Internals

LLM Name	Toffee 3B Beta
Repository 🤗	https://huggingface.co/maxzt/toffee-3b-beta
Base Model(s)	Ministral 3 3B Base 2512 mistralai/Ministral-3-3B-Base-2512
Model Size	3b
Required VRAM	7.7 GB
Updated	2026-08-15
Maintainer	maxzt
Model Type	mistral
Instruction-Based	Yes
Model Files	7.7 GB
Supported Languages	en
Model Architecture	MistralForCausalLM
License	mit
Context Length	8192
Model Max Length	8192
Transformers Version	5.5.0
Tokenizer Class	TokenizersBackend
Padding Token	<pad>
Vocabulary Size	131072

Rank the Toffee 3B Beta 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 Toffee 3B Beta

Best Alternatives	Context / RAM	Downloads	Likes
Ministral 3B Instruct	128K / 6.7 GB	7607	86
Mistral 3B Instruct V0.2	32K / 11.5 GB	4	4
...inerva 3B Llama3 Instruct V0.1	16K / 5.8 GB	1252	0
Minerva 3B Instruct V1.0	16K / 5.8 GB	19	7
Minerva 3B Instruct V1.0	16K / 5.8 GB	12	7
Phi 3 Phituguese 3B FP16	4K / 7.6 GB	7	0
Mistral 3B Instruct V0.2 Init	32K / 5.7 GB	127	3