Our own model
M1 is Stellar’s large-model business, developing language and task-processing capabilities.
STELLAR / M1 MODEL & API
M1. Intelligence, inside your product.
Stellar’s own large model, delivered through an API for developers, businesses and long-term partners. Our research aims to help software understand requests and assist with the work that follows.
API services
Model capabilities are delivered through an interface for use in your own products.
Long-term technology supply
Continue model and technology collaboration with existing partners, with agreed integration and support.
M1 / IN YOUR APPLICATION
From a request to the next step.
What matters is how a model fits into the work. These application flows explain the input, the model’s role and how results return to your system.
INPUT
A user describes a problem. The application supplies relevant product or page information.
M1 / API
Interpret the question and organize an answer or instructions from the supplied material.
APPLICATION
The application presents the answer. Account actions or decisions can continue through the appropriate workflow.
Application value
Put assistance where users need it and shorten the path to useful information.
Illustrative application flowINPUT
A team supplies documents, a task and material to organize.
M1 / API
Work on the specified task and produce a result for the next stage.
APPLICATION
The business system receives the result for review or continuation in an existing workflow.
Application value
Bring model assistance to information processing and reduce repetitive reading and organization.
Illustrative application flowINPUT
An application submits a task, context and output requirements through the API.
M1 / API
The model produces a response, returned through the interface.
APPLICATION
Developers use the result in their product logic, evaluate it and refine the integration.
Application value
Add model services while retaining control over the interface, workflow and experience.
Illustrative application flowM1 / CAPABILITIES & ACCESS
The model. The interface.
M1 provides model capabilities. The API platform provides access and management. Evaluate task fit, interface support and actual results together.
- 01Models & versions
- Call the model using its platform identifier. Current versions and model-specific capabilities are listed in the console.
- 02API access
- The API platform provides an OpenAI-compatible interface, making familiar request patterns available.
- 03Streaming & output
- The platform offers streaming, tool calls and structured JSON options. Support varies by model; refer to platform documentation.
- 04Usage & management
- Keys, usage, balance and billing management help take an integration from testing to ongoing use.
Test with your own tasks, then choose the model, output method and review process that fit.
DEVELOPERS / GET STARTED
Start with one request.
Get your key and connection details from the model platform. Test a small task before integrating it into your software.
- 01
Create an account
Sign in to the API platform and review available models and billing information.
- 02
Get connection details
Obtain an API key, Base URL and the actual model ID from the console.
- 03
Test a task
Submit a question with context. Inspect the response format and task results.
- 04
Connect the workflow
Plan error handling, usage management and any necessary human review.
Request structure / JavaScript
// Values supplied by your API console.
const baseUrl = 'YOUR_BASE_URL';
const modelId = 'YOUR_MODEL_ID';
const apiKey = 'YOUR_API_KEY';
const response = await fetch(
baseUrl.replace(/\/$/, '') + '/chat/completions',
{
method: 'POST',
headers: {
'Content-Type': 'application/json',
Authorization: `Bearer ${apiKey}`,
},
body: JSON.stringify({
model: modelId,
messages: [{role: 'user', content: 'Hello'}],
}),
},
);
if (!response.ok) throw new Error(String(response.status));
const result = await response.json();Run this on your server. Replace the Base URL, key and model ID with console values. This example shows the request structure.
MODEL / OPERATIONS
Built into a product. Ready for daily use.
Results & cost
Evaluate response quality, wait time and usage on real tasks. Estimate cost using current platform pricing.
Data & control
Review service data policies before integrating. Decide what to submit, when to review and how to handle exceptions.
Versions & support
Follow model and interface updates. Agree integration scope and support arrangements for long-term partnerships.
M1 / RESEARCH DIRECTION
From answers to reliable collaboration.
We want software to do more than present information: to understand requests and help with the work. Research follows real applications so each improvement has a purpose that can be evaluated.
- 01 / RESEARCH DIRECTION
Understand tasks more reliably
Develop request understanding and output consistency across different wording and contexts.
- 02 / RESEARCH DIRECTION
Connect with systems more naturally
Study how model results fit into business workflows, making capabilities easier to integrate into useful features.
- 03 / RESEARCH DIRECTION
Improve through evaluation
Use evaluation from product and partner scenarios to advance capabilities, operational efficiency and integration methods.
LET’S BUILD / M1
Your applications. A lasting partnership.
M1 is a model and a continuing technology business. We maintain model and foundational technology collaboration with existing long-term partners and welcome specific application and research needs.
- 01Describe your application and existing systems
- 02Evaluate model fit and API integration together
- 03Agree scope, support and future iteration
QUESTIONS / ANSWERS
Questions, answered.
How are M1 and AskAIs different?
M1 is the model and API technology business for developers and partners integrating AI into their own systems. AskAIs is a ready-to-use AI support product: AI answers first and teammates take over when needed. Current AskAIs features and model options are listed in the product.
How do I start using the API?
Create an account on the model platform and obtain a key, Base URL and actual model ID. Follow platform integration instructions and evaluate results on your own tasks.
Where are prices, limits and data policies?
Refer to the API platform for current model prices, versions, calling limits and service policies. Contact the team to discuss technology supply and support for long-term partnerships.
Will existing partnerships continue?
Yes. Models and foundational technology supply remain part of Stellar’s business. Existing partnerships continue under agreed arrangements.
Technology & business updates
Model API platform
Find available models, integration instructions and billing information.
AskAIs product
Explore AI replies, team handoff and knowledge-based support.