What are the hardware requirements for MiniMax M2.5 Bf16 Heretic?

MiniMax M2.5 Bf16 Heretic requires approximately 440 GB of VRAM and supports a context window of 192K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed MiniMax M2.5 Bf16 Heretic and how large is it?

MiniMax M2.5 Bf16 Heretic is developed by RadicalNotionAI, a model with 228.7b 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 MiniMax M2.5 Bf16 Heretic?

MiniMax M2.5 Bf16 Heretic is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

MiniMax M2.5 Bf16 Heretic by RadicalNotionAI — VRAM 440GB, 192K context

Name: MiniMax M2.5 Bf16 Heretic
Author: RadicalNotionAI

MiniMax M2.5 Bf16 Heretic is an open-source language model by RadicalNotionAI. Features: 228.7b LLM, VRAM: 440GB, Context: 192K, License: other, LLM Explorer Score: 0.23.

Abliterated Ara Conversational Custom code Decensored Endpoints compatible Heretic Minimax m2 Region:us Safetensors Sharded Tensorflow Uncensored

Model Card on HF 🤗: https://huggingface.co/RadicalNotionAI/MiniMax-M2.5-bf16-heretic

MiniMax M2.5 Bf16 Heretic Parameters and Internals

LLM Name	MiniMax M2.5 Bf16 Heretic
Repository 🤗	https://huggingface.co/RadicalNotionAI/MiniMax-M2.5-bf16-heretic
Model Size	228.7b
Required VRAM	440 GB
Updated	2026-07-17
Maintainer	RadicalNotionAI
Model Type	minimax_m2
Model Files	10.0 GB: 1-of-46 10.0 GB: 2-of-46 10.0 GB: 3-of-46 10.0 GB: 4-of-46 10.0 GB: 5-of-46 10.0 GB: 6-of-46 10.0 GB: 7-of-46 10.0 GB: 8-of-46 10.0 GB: 9-of-46 10.0 GB: 10-of-46 10.0 GB: 11-of-46 10.0 GB: 12-of-46 10.0 GB: 13-of-46 10.0 GB: 14-of-46 10.0 GB: 15-of-46 10.0 GB: 16-of-46 10.0 GB: 17-of-46 10.0 GB: 18-of-46 10.0 GB: 19-of-46 10.0 GB: 20-of-46 10.0 GB: 21-of-46 10.0 GB: 22-of-46 10.0 GB: 23-of-46 10.0 GB: 24-of-46 10.0 GB: 25-of-46 10.0 GB: 26-of-46 10.0 GB: 27-of-46 10.0 GB: 28-of-46 10.0 GB: 29-of-46 10.0 GB: 30-of-46 10.0 GB: 31-of-46 10.0 GB: 32-of-46 10.0 GB: 33-of-46 10.0 GB: 34-of-46 10.0 GB: 35-of-46 10.0 GB: 36-of-46 10.0 GB: 37-of-46 10.0 GB: 38-of-46 10.0 GB: 39-of-46 10.0 GB: 40-of-46 10.0 GB: 41-of-46 10.0 GB: 42-of-46 10.0 GB: 43-of-46 10.0 GB: 44-of-46
Model Architecture	MiniMaxM2ForCausalLM
License	other
Context Length	196608
Model Max Length	196608
Transformers Version	5.5.0
Vocabulary Size	200064

Rank the MiniMax M2.5 Bf16 Heretic 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 MiniMax M2.5 Bf16 Heretic

Best Alternatives	Context / RAM	Downloads	Likes
MiniMax M2.7	200K / 75.2 GB	923544	1235
List 3.0 Ultra Coder Brain	200K / 75.2 GB	22	10
MiniMax M2.7 BF16	200K / 147.3 GB	18	0
MiniMax M2	192K / 77.7 GB	103020	1492
MiniMax M2.5	192K / 74 GB	772099	1502
MiniMax M2.7	192K / 213 GB	247	2
MiniMax M2.7	192K / 75.2 GB	58	18
MiniMax M2.5	192K / 75.2 GB	19	5
MiniMax M2.1	192K / 75.2 GB	54	13
MiniMax M2	192K / 77.7 GB	24	17

Note: green Score (e.g. "73.2") means that the model is better than RadicalNotionAI/MiniMax-M2.5-bf16-heretic.