Nex-N2.6
A multimodal agentic model based on GLM-5.3-Flash for computer use, programming and long tasks with tools. It is published in FP8, about 321B parameters. In this version the authors reduced repetitive reasoning and improved adherence to the tools available in the environment. On the OSWorld test it scores 83.0 against 82.2 for the previous Nex-N2.5-Pro and 78.0 for the base GLM-5.3-Flash, and on Terminal-Bench 2.1 it scores 84.1 against 84.3 for the base. The figures come from the authors. It needs a cluster of several server graphics cards.
- Developer
- Nex-AGI, not stated
- First release
- Nov 2026
- Latest release
- Nov 2026
- Sizes
- 321B (FP8)
- License
- Commercial use allowedApache 2.0 (based on GLM-5.3-Flash)
- Russian
- Not stated
- Running
- On your own serverNeeds a GPU
- Industries
- Software development, Customer support, Finance
What it does
- an agent that operates a computer through screenshots
- an agent for programming and the terminal
- long tasks with tool calls
Where it is used
Hardware requirements
Versions
- Nex-N2.6 (FP8)
- Nex-N2.5-Pro
How to run it
I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.
Frequently asked questions
Can Nex-N2.6 be used in a commercial project?
Yes. License: Apache 2.0 (based on GLM-5.3-Flash). It allows commercial use, but it is still worth having a lawyer review the license before launch.
What hardware does Nex-N2.6 need?
At minimum: Server with several GPUs — flagship versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.
Does Nex-N2.6 support Russian?
The model card does not list languages, so Russian support cannot be promised — it has to be tested on your own examples.
Where can I download Nex-N2.6 and what does it cost?
The Nex-N2.6 weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.
How I deploy it for clients
- SelectionI pick the model size for your task and hardware and test it on your examples.
- DeploymentI deploy it on your server or in a closed network and provide an API.
- Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
- IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.
Similar models
One of the oldest Chinese open lines: from ChatGLM-6B to GLM-5.3. Strong at agentic tasks and programming; GLM-5.3-Flash understands images and is released under MIT.
DetailsImage + textMage-VL-CoworkMicrosoft · USACommercial use allowedA compact 4B model that operates a computer by itself: it looks at a screenshot and clicks, and when that is more efficient it switches to the command line and edits files with code. Both roles are in one model, with no external orchestrator. On the OSWorld-Verified test (50 steps) it scores 66.0%, which the authors say is the best result among 4B models, and 87.5% on operating-system tasks. It was trained from the intermediate Mage-VL in five stages, with a set of almost 294 thousand synthetic tasks over 6,915 skills. The context for training and running is 32,768 tokens, and the architecture holds up to 262,144. It suits office automation and assistants on your own computer. The figures come from the authors.
DetailsTextQwenAlibaba · ChinaCommercial use with conditionsA family of language models with strong Russian language support, from small versions for a laptop to a flagship on par with commercial APIs.
DetailsSource: huggingface.co/nex-agi/Nex-N2.6. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.


