BetaFree access while in Beta
Scientific models as a service
Run leading AI models for biology and chemistry without managing GPUs or infrastructure.
Access is free for the duration of the Beta. Requests are reviewed manually; commercial terms may follow later with notice.
Valuable models are often locked behind hardware, packaging and operations work. We take that work off your plate so the model is simply there when you need it.
Structure prediction, protein design, genomics, small-molecule generation, single-cell analysis and imaging models share one access pattern, one key and one billing vocabulary.
Designed to take you from a handful of requests to campaigns and cohorts, with managed capacity that can grow with your work.
You never source accelerators, provision servers, predict demand, tune autoscaling or babysit servers. You send inputs and get scientific outputs back.
If the model you rely on is not listed, tell us. Nebius can add and operate it so your team keeps its focus on the science.
Fold, dock, design and annotate at the pace of your programme. Onboard new hires with a single key instead of an infrastructure tutorial, and keep sensitive infrastructure decisions out of the critical path.
Use the same models that appear in the papers you read, without a grant for GPUs or a week of environment setup. Ask your AI assistant to run a model through MCP and stay in the flow of your analysis.
Tell us who you are and what you want to run. It takes two minutes.
The Nebius team reviews your request personally, agrees the beta participation conditions with you and emails your API key.
Pick your client, click Copy MCP, paste, and start asking your assistant to run models.
Choose your client and copy a ready-to-paste configuration. Replace the placeholder with the key you receive by email, and your assistant can call the models as tools.
{
"mcpServers": {
"nebius-scientific-ai": {
"type": "http",
"url": "https://89.169.99.188/mcp",
"headers": {
"Authorization": "Bearer ${NEBIUS_SCIENTIFIC_AI_API_KEY}"
}
}
}
}{
"mcpServers": {
"nebius-scientific-ai": {
"url": "https://89.169.99.188/mcp",
"headers": {
"Authorization": "Bearer ${env:NEBIUS_SCIENTIFIC_AI_API_KEY}"
}
}
}
}{
"inputs": [
{
"type": "promptString",
"id": "nebius-scientific-ai-api-key",
"description": "Nebius Scientific AI Platform API key (sent to you by email)",
"password": true
}
],
"servers": {
"nebius-scientific-ai": {
"type": "http",
"url": "https://89.169.99.188/mcp",
"headers": {
"Authorization": "Bearer ${input:nebius-scientific-ai-api-key}"
}
}
}
}[mcp_servers.nebius_scientific_ai] url = "https://89.169.99.188/mcp" bearer_token_env_var = "NEBIUS_SCIENTIFIC_AI_API_KEY"
https://89.169.99.188/mcp
{
"mcpServers": {
"nebius-scientific-ai": {
"transport": "streamable-http",
"url": "https://89.169.99.188/mcp",
"headers": {
"Authorization": "Bearer NEBIUS_SCIENTIFIC_AI_API_KEY"
}
}
}
}Replace NEBIUS_SCIENTIFIC_AI_API_KEY with the key you receive by email. Keys are never embedded on this page.
Fill in the Request Beta access form. The Nebius team reviews every request and replies by email. Access is not automatic, and we may prioritise requests to keep the Beta responsive for everyone.
After your request is approved, we agree the available models, capacity and beta participation conditions with you. Keys are then created and emailed manually. We never publish keys on this site, and you should never paste a key into a public place.
Yes, while the service is in Beta every listed model is billed at $0.00. Any future paid offering will have separate pricing and terms, communicated before you choose to participate.
The Models page polls our live catalog and marks each model as Available now, Preview or Coming soon. Access for your account, available capacity and applicable model licence terms are confirmed during onboarding. If the catalog cannot be reached the page says so instead of guessing.
Beta capacity is shared and bounded. Billing units describe how usage is measured; request limits vary by model. We do not offer an SLA during Beta; if you need committed throughput, tell us in your request and we will discuss options.
The Model Context Protocol lets AI assistants and coding agents call tools over a standard interface. Our MCP endpoint exposes the scientific models as tools, so you can ask an assistant to fold a sequence or dock a ligand without writing integration code.
Often, yes. Use the Host your model form to tell us about the model, its licence and expected usage. We will review feasibility and come back with a proposal. We do not accept confidential weights or sensitive data through the form itself.
We recommend starting with public, synthetic or non-sensitive research data. If your evaluation requires personal data, patient information or confidential datasets, discuss this with us during onboarding before uploading them. Authorised platform operators can access uploaded data and request logs for support and troubleshooting.
The beta supports scientific research and evaluation, including molecular dynamics and drug discovery. Results require independent scientific validation. It is not intended for clinical diagnosis or treatment decisions.