Your AI agents keep working, with no session limits.
Private pilot for One Mount Group and Techcombank teams.
Managed AI platform for engineering teams in Vietnam and Southeast Asia.
Platform and self-hosted models are live in pilot today, alongside Nexus Composer — a desktop app that lets your coworking agents call our Platform models. Semantic routing, elastic GPU capacity, and more agent tools come next.
- Evaluating today
- Engineering teams at One Mount Group and Techcombank.
- Available in pilot
- Platform models, self-hosted models, and Nexus Composer.
- Private preview
- AI training framework.
01
Target teams
Who OneNexus is for
Power your products and internal tools through one standard API.
Train models on your own approved data, watch progress, and put the results to work.
02
Platform models
One API. Hosted models. Clear controls.
Call models from the shared catalog with a Platform API key. No GPU quota required.
OpenAI-compatible API
Works with your existing OpenAI setup.
Explicit model selection
The model you specify is the model that runs.
Per-company access and quotas
Issue keys, track usage, and enforce limits for your company.
Usage visibility
Review request and token usage by model in the Console.
Integrate in minutes
Change the base URL, keep your code.
python
import os
from openai import OpenAI
client = OpenAI(
base_url="https://catalog.onenexus-do.cloud/v1",
api_key=os.environ["ONX_API_KEY"],
)
resp = client.chat.completions.create(
model="glm-5.3",
messages=[{"role": "user", "content": "Refactor this module"}],
)
print(resp.choices[0].message.content) 03
Products
Bring your models. Connect AI agents. Train on your data.
Available in pilot
Self-hosted models
Deploy a model your organisation owns and call it through your own private endpoint.
Self-service on the platform. Inference GPU quota required.
Available in pilot
Nexus Composer
Call OneNexus-hosted models from the agents you already use.
Desktop app: macOS available now, Windows and Ubuntu coming soon.
Private preview
OneNexus Training
Upload your data, train a model, watch progress, and put the result to work.
Self-service on the platform. Training GPU quota required.
04
About OneNexus
Vietnam’s AI Infrastructure & Platform Company
We are building the full AI value chain — AI cloud, data platform, and AI applications — for enterprises in Vietnam and Southeast Asia. Today, our managed AI infrastructure is live in pilot. The data platform and enterprise AI applications follow on our roadmap.
Get started
Request pilot access
Tell us what you want to build and we’ll help you choose the right starting point.
Who qualifies, what it costs, how your data is handled, and what happens after you submit — all on Pilot & Security.
Not part of either One Mount Group or Techcombank? Choose “External interest” below — we’ll notify you when access opens.