What are the hardware requirements for Nex N2.5 Max?

Nex N2.5 Max requires approximately 115.4 GB of VRAM and supports a context window of 1024K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Nex N2.5 Max and how large is it?

Nex N2.5 Max is developed by nex-agi, a model with 1600.8b 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 Nex N2.5 Max?

Nex N2.5 Max is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Nex N2.5 Max by nex-agi — VRAM 115.4GB, 1024K context

Name: Nex N2.5 Max
Author: nex-agi

Nex N2.5 Max is an open-source language model by nex-agi. Features: 1600.8b LLM, VRAM: 115.4GB, Context: 1024K, License: apache-2.0, LLM Explorer Score: 0.44.

Conversational Deepseek v4 Endpoints compatible Fp8 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/nex-agi/Nex-N2.5-Max

Nex N2.5 Max Parameters and Internals

LLM Name	Nex N2.5 Max
Repository 🤗	https://huggingface.co/nex-agi/Nex-N2.5-Max
Model Size	1600.8b
Required VRAM	115.4 GB
Updated	2026-09-11
Maintainer	nex-agi
Model Type	deepseek_v4
Model Files	3.8 GB: 1-of-644 2.5 GB: 2-of-644 2.5 GB: 3-of-644 2.5 GB: 4-of-644 2.5 GB: 5-of-644 2.5 GB: 6-of-644 2.5 GB: 7-of-644 2.5 GB: 8-of-644 2.5 GB: 9-of-644 2.5 GB: 10-of-644 2.9 GB: 11-of-644 2.5 GB: 12-of-644 2.5 GB: 13-of-644 2.5 GB: 14-of-644 2.5 GB: 15-of-644 2.5 GB: 16-of-644 2.5 GB: 17-of-644 2.5 GB: 18-of-644 2.5 GB: 19-of-644 2.5 GB: 20-of-644 2.5 GB: 21-of-644 2.9 GB: 22-of-644 2.5 GB: 23-of-644 2.5 GB: 24-of-644 2.5 GB: 25-of-644 2.5 GB: 26-of-644 2.5 GB: 27-of-644 2.5 GB: 28-of-644 2.5 GB: 29-of-644 2.5 GB: 30-of-644 2.5 GB: 31-of-644 2.9 GB: 32-of-644 2.5 GB: 33-of-644 2.5 GB: 34-of-644 2.5 GB: 35-of-644 2.5 GB: 36-of-644 2.5 GB: 37-of-644 2.5 GB: 38-of-644 2.5 GB: 39-of-644 2.5 GB: 40-of-644 2.5 GB: 41-of-644 2.9 GB: 42-of-644 2.5 GB: 43-of-644 2.5 GB: 44-of-644 2.5 GB: 45-of-644
Model Architecture	DeepseekV4ForCausalLM
License	apache-2.0
Context Length	1048576
Model Max Length	1048576
Transformers Version	5.8.0
Vocabulary Size	129280
Torch Data Type	bfloat16

Rank the Nex N2.5 Max 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