AI MODELS

Every leading AI model. One governed platform. No lock-in.

Give your users access to frontier models from OpenAI, Anthropic, Gemini, and more without rebuilding your governance framework. In ONEchat, model flexibility is connected to real work: users can select the right model for a task while administrators decide which models are available, where they can be used, and how usage is governed.

See ONEchat in action

Model picker with ChatGPT, Claude, Gemini and Grok above a governed chat window

Supported models grid

Last updated date: September 2026

ChatGPT, Claude, Gemini and Grok available in nebulaONE

OpenAI models:

Up to ChatGPT 5.6

Anthropic models:

Up to Claude Fable 5.1, Opus 5, Sonnet 5

Google models:

Up to Gemini 3.6

Grok

Up to Grok 4.3

Additional models from:

Microsoft, Meta, Kimi, Mistral, and more

3 STEPS

Model addition timeline process

1. Evaluation

When a significant new model is released by a leading provider, our team evaluates it against performance, safety, and Azure AI Foundry compatibility criteria. Most major releases are assessed within days of public availability, so your institution is never waiting long to access the latest capabilities.

2. Integration

Once a model clears evaluation, it is integrated into the nebulaONE® platform and made available to administrators to decide whether to offer it to their users.

3. Deployment

Administrators decide which models to deploy through their Foundry accounts. Once deployed, you can decide whether the model is available for agents or ONEchat.

Task-to-model decision matrix

Use case / task type

Model examples

Rationale

Cost impact

Complex reasoning & synthesis

OpenAI ChatGPT 5.6 Sol Claude Fable 5.1, Opus 5

Best-in-class multi-step reasoning for research synthesis, policy analysis, and complex document review.

High / $$$

High-volume search & retrieval

OpenAI ChatGPT 5.6 Terra Claude Sonnet 5 Gemini 3.6

Fast, cost-efficient performance for document Q&A, knowledge base search, and high-frequency student-facing queries.

Low–moderate / $$

Data extraction & classification

OpenAI ChatGPT 5.6 Luna Claude Haiku 4.5 Gemini 3.6 Flash

Accurate structured output for form processing, content tagging, and data normalization at scale.

Low / $

Not every task requires your most powerful or most expensive model. nebulaONE makes it straightforward to offer the right model mix for your users to complete their tasks while keeping your costs under control. Administrators set user and group usage limits centrally, giving your institution full visibility and control over AI spend before it becomes a board-level conversation. This is the core of nebulaONE’s consumption-based pricing model: you pay for what your institution actually uses, not for seats that sit idle.

See how consumption-based pricing works

RESOURCES