M MODEL ACCESS, MADE FOR TEAMS

More possibilities.
More control.

Bring Claude, GPT and Gemini into your workflow. Give every project its own keys, every team a budget, and every decision a clearer view of usage.

One API origin. A workspace that belongs to your team.

YOUR APPLICATIONSCONCEPT VIEW
AI agentsYour productInternal tools
model100api.model100.org
API
✳Claude
GGPT
✦Gemini
A practical foundation for your AI stack.Multiple model familiesOrganization workspacesToken-level metering

A WORKSPACE, NOT JUST AN API KEY

Keep the whole team
in the picture.

Move from shared credentials and scattered spreadsheets to a clear structure for people, projects and model usage.

E Example workspaceINTERACTIVE PREVIEW · SAMPLE DATA
⌁

Your people. Your projects. Your organization’s data.

Ownership, from day one.

Admins manage members. Members create project keys. Everyone sees the usage they’re permitted to see.

Budgets with boundaries.

Share an organization balance and set spending limits for individual members and API keys.

Numbers you can work with.

Review model and project spending, cache hit rates, and minute-level usage reports through your Console and APIs.

DESIGNED TO FIT YOUR WORKFLOW

A familiar API.
A clearer operating picture.

Use the API format supported by your chosen model. Create a key for its enabled platform, connect your application, and track usage in your team’s Console.

/v1/messagesClaude
/v1/responsesGPT
/v1/modelsAvailable models for your key

Supported models and platform access depend on your workspace configuration. Request formats and features can differ by model.

Discover your available models
curl 'https://api.model100.org/v1/models' \
  -H 'Authorization: Bearer YOUR_API_KEY'
{ }

Bring monitoring into your own tools.

Generate a read-only monitoring token in Console. Query current concurrency and RPM without managing a login cookie in your scripts.

ORGANIZATION-SCOPED ACCESS

KNOW WHAT YOU’RE USING

A clear view of cost.
Before you scale.

Review the model prices and effective rates configured for your workspace, then follow actual usage by project, member and key.

◫

Model pricing, visible

Your Console shows standard prices and your applicable workspace rates for supported catalog models.

≋

Caching, explained

See cache reads and writes alongside other input and output tokens. Understand how much of your input was served from cache.

◎

A plan for your workload

Discuss model access, pricing and organization concurrency with our team before onboarding.

FROM EVALUATION TO EVERYDAY USE

Start with your use case.

We help configure your workspace. Your team takes it from there.

ALIGN

Tell us what you’re building

Share the model families, expected workload and team structure you need.

CONFIGURE

Get your own workspace

We provision your organization, administrator access, model permissions and agreed limits.

CONNECT

Create keys. Run your tests.

Invite your team by creating member accounts, organize projects and evaluate the API with your own workloads.

A FEW USEFUL DETAILS

Before you get started.

Which models can my team use?

Model100 supports access to Claude, GPT and Gemini model families. Exact model availability and platform permissions are configured for each workspace. Use your Console and GET /v1/models with the relevant API key to check what is available to you.

Can everyone on my team create their own keys?

Your organization administrator creates member accounts. Members can then create projects and API keys within their accounts. Admins can review organization usage and manage member limits; members see their own permitted data.

How do billing and top-ups work?

Members share an organization balance. Top-ups are currently arranged with the Model100 team and credited manually. Your admin can set member spending limits, and individual keys can also have spending limits. Usage-based limits are not a guarantee against overspend from requests already in flight.

Does available concurrency guarantee a request will run?

No. Concurrency reports are approximate observations, not capacity reservations. Platform ceilings share the organization limit, and balance, model availability and upstream limits can still affect a request. RPM measures recent traffic; it is not an RPM allowance.

Is this the customer login page?

This is the Model100 product website. Each organization receives its own Console address during onboarding. If you already have an account, use the workspace link provided by your administrator.

Is Model100 an official model-provider website?

Model100 is an independent API access service. Claude, GPT and Gemini are names associated with their respective providers. Model and feature availability should be confirmed for your workspace.