Model library / nex-agi
Nex-N2.5-Max is a 1.6-trillion-parameter text-only MoE model designed for long-horizon agentic tasks.
6 of 384 experts active per token
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
Nex-N2.5-Max is a deepseek_v4 causal language model released by Nex-AGI. The publisher describes it as a next-generation agentic model for long-horizon tasks in real-world environments, with capabilities for continuous operation, self-correction via visual feedback, and autonomous computer and browser interaction. It is part of the Nex-N2.5 family that includes mini and Pro variants.
In practice the model comprises 1,600.8 billion parameters organised as a mixture-of-experts with 384 experts, of which six are active per token. It supports a context window of 1,048,576 tokens, uses fp8 weight format, and is released under the Apache-2.0 licence. Deployers must provision sufficient memory and compute to handle the MoE routing and the very long context length.
Specification
| Repository | nex-agi/Nex-N2.5-Max |
| Publisher | nex-agi |
| Published | 7 September 2026 |
| Architecture | DeepseekV4ForCausalLM |
| Model type | deepseek_v4 |
| Parameters | 1,600,787,478,430 |
| Experts | 384 total, 6 active per token |
| Layers | 61 |
| Hidden size | 7,168 |
| Attention heads | 128, 1 key/value heads |
| Context length | 1,048,576 tokens |
| Vocabulary | 129,280 tokens |
| Quantisation | fp8 |
| Licence | apache-2.0 |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 2981.7 GiB | Full precision as released |
| fp8 | 1490.8 GiB | 8-bit, near-lossless on most models |
| int4 | 745.4 GiB | 4-bit, smallest footprint |
The cache costs 122 KiB per token at 16-bit, so the full 1,048,576-token context of one request needs about 122.0 GiB. Concurrency multiplies that number, not the weights.
That figure is an upper bound: this model uses sliding-window attention with a 128-token window on some layers, so long requests cache less than this.
Hardware
At the released precision, with 90% of the card given to the server and 2.5 GiB kept for the runtime. The context column is what is left for the cache on that setup.
| GPU | Cards needed | Free for cache | Context it holds |
|---|---|---|---|
| RTX 4090 24GB | more than 8 | — | — |
| RTX 5090 32GB | more than 8 | — | — |
| L40S 48GB | more than 8 | — | — |
| A100 80GB | more than 8 | — | — |
| H100 80GB | more than 8 | — | — |
| RTX PRO 6000 96GB | more than 8 | — | — |
| H200 141GB | more than 8 | — | — |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve nex-agi/Nex-N2.5-Max \ --tensor-parallel-size 8 \ --max-model-len 1048576 \ --gpu-memory-utilization 0.90
At the released precision the weights do not fit on eight of the largest cards above, so this needs more than one node, or a quantised build.
Questions
The weights take about 1541 GiB at the released precision. No single card in the table above holds that, so it needs several GPUs or a lower precision.
The repository declares apache-2.0. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 1,048,576 tokens. Whether the whole window is usable depends on the memory left for the cache, which the hardware table works out per card.
This page is a reference, and Aiqre does not serve this model today. We deploy open-weight models on dedicated EU hardware on request, so if you want an OpenAI-compatible endpoint for it, email [email protected] with the model name and roughly what volume you expect.
Also in the library
Nex-N2.5-Pro is a 396.8 billion-parameter MoE language model designed for long-horizon, agentic tasks with multimodal and compute…
Nex-N2.5-mini is a 35.1 billion-parameter mixture-of-experts language model released under Apache-2.0.
DeepSeek-V4-Pro-0813 is a 1650.5B mixture-of-experts open-weight model published by deepseek-ai on Hugging Face. It uses the Deep…
Qwen3.8-2.4T-A95B is a 2.4 trillion-parameter causal language model with 95 billion active parameters, supporting up to 262 k tok…
Hosting
Tell us the model and roughly what volume you expect. We reply with a price and an OpenAI-compatible endpoint, with your prompts never stored.