Can we keep our existing providers, and do we have to use OpenCode's inference?

You can connect your existing providers or a custom gateway. Console supports more than 30 providers, so your team can keep using the models and provider accounts it already has. OpenCode's inference is optional.

Which identity providers are supported?

Connect your workspace to Okta, Microsoft Entra ID, or Google Workspace. Members sign in with the identity provider your organization already uses.

We need a BAA or custom terms, but we are a small team. What are our options?

Contact our team with your requirements. We can discuss your team's needs, any required agreements, and the enterprise options that may fit your organization.

Everything you need to work with AI

High-quality inference for frontier and open-weight models, low-cost subscriptions with generous limits, and the tools your team needs to manage spend and collaborate effectively

Get started

Inference

Frontier and open-weight models

Access every model on the planet in a single interface through the world's largest inference provider. Serving 600 trillion tokens a month.

Learn more

Model	Features	Input	Output	Cache read	Cache write
Claude Opus 5.5	+3	$4.00	$20.00	$0.20	$5.00
Claude Sonnet 5.5	+3	$2.00	$10.00	$0.20	$2.50
Nemotron 3 Ultra		$0.60	$2.40	$0.12	-
GPT-5.6 Terra	+3	$2.00	$12.00	$0.20	$2.50
Gemini 3.8 Flash	+5	$1.50	$7.50	$0.15	-
Mistral Large 4	+2	$0.68	$2.09	$0.07	-
GPT-6 Astra	+4	$10.00	$50.00	$1.00	$12.50

NVIDIA3 modelsConnectConnectingConnected

Anthropic17 modelsConnect

An example catalog shows model capabilities, input and output token prices, cache prices, and enabled access. NVIDIA connects and its Nemotron 3 Ultra model comes online in the catalog, and Anthropic offers 17 models to connect next.

Subscriptions

Monthly subscription packages

OpenCode Go is a subscription that gives you and your team access to the best open-weight models at a fixed monthly cost.

View plans

Estimated requests per five hours and monthly usage for eleven Go models, matching the usage limits in the Go documentation. Yellow bars represent Go and orange bars represent Go Plus. For example, DeepSeek V4.1 Flash allows about 43,330 requests and $100 of monthly usage on Go, and 86,670 requests and $200 on Go Plus. The update time is illustrative.

Teams

Manage agents across your team

Connect your own providers, set up single sign-on, set budgets and control model access for your whole team in a central place. Connect integrations to give your agents secure access to useful tools like Slack or Email.

Learn more

Aalan@anoma.ly

Member

Pending·Expires Nov 2

Ffilip@anoma.ly

Member

Pending·Expires Nov 5

aarav@anoma.ly

MemberAdmin

Ludvig

Oct 21

dax@anoma.ly

Admin

Sep 14

frank@anoma.ly

Owner

Oct 3

Recurring creditsequals$0useOne-time credits

When balance falls below$1,000add

$500

**6492

Allow

workspace spend to reach$10,000✓every month

Allow each

user to spend$21,000✓every month

Example workspace with pending invitations and Member, Admin, and Owner roles. Spending controls combine recurring and one-time credits, automatically reload $500 below a $1,000 balance, and set monthly limits of $10,000 for the workspace and $1,000 for each user.

Enterprise

Talk to our team

For volume pricing, inference commitments, security reviews, BAAs and custom contracts, or deployment needs like private networks and existing gateways.

Explore enterprise

Frequently asked questions