Pricing

Workflows is a no code design builder. You drag model blocks onto a canvas, connect them, choose a few options, then press Execute. We run the GPUs, scale out jobs, and keep everything in order so you do not need to write code or manage servers.

You can go from structure prediction to design and scoring in one place. For a full list of models that plug into Workflows, see the platform library.

No. If you can click and drag, you can create a workflow. Built in help bubbles and guides teach you step by step. Power users can run many jobs in parallel. Newcomers can stay with defaults that just work.

You can change plan any time from the Pricing page or from your account. Upgrade when you need more credits or higher parallel jobs, or move back down when you are running fewer jobs.

Credits are the currency that pays for compute. Each job uses credits based on the model and the size of the problem you run.

For reference, 1 credit = 5.4 minutes of GPU time. Larger systems and heavier models simply use more credits.

In Workflows, each node you execute counts as one job. A workflow with 5 nodes can run up to 5 jobs, and each job draws from your credit balance based on its model and input size.

Total credits used is the sum of all jobs run in your workflow. Parallel nodes let you fan out work and finish faster, while still tracking credit usage clearly.

For the Free tier, credits refill on the 1st of each month at 00:00 (UTC).

For paid monthly plans, unused credits roll over each month. New credits are added monthly on the same calendar day that you started the plan.

Yearly plans receive the full year of credits up front on the purchase date. Your next yearly refill happens exactly one year after that purchase date.

If you hit your limit sooner than expected you can upgrade your tier at any time to increase credits and parallel jobs. You can always see your current balance on your account page.

Admin: the person who purchases the Team plan becomes the team admin. From Account → Team in your account you can copy your invite link, add or remove members, and manage roles.

Pooled credits: all team members draw from a shared credit pool. This is ideal for collaborative sprints and shared budgets.

Parallelism: teams can execute up to 192 concurrent jobs (subject to plan limits). Use Workflows parallel blocks to fan out workloads and move faster as a group.

A job is a single execution. For example, run a Chai-1 fold, score with DockQ, or design with AntiBMPNN. Each node in Workflows or each run on an individual model page counts as 1 job.

Jobs often run in parallel up to your plan concurrency limit. More parallel jobs means faster results, and credits are charged for each job that runs.

Yes. You can manage billing from the Pricing page or from your account. You can cancel, upgrade, or downgrade whenever your usage changes.

We use proration for plan changes, so your limits and credit balance update shortly after you switch.

Your data and IP stay yours. We do not train models on your jobs. Projects are private by default and you can export results at any time.

We are GDPR compliant and support OIDC/SSO. You control your data from your dashboard and can delete it at any time. Deletion is permanent, so please double check before removing anything.

1) Create an account on the sign up page. 2) Open a template or drop nodes onto the canvas in Workflows. 3) Click Execute to run.

You can upgrade when you need more credits, higher parallelism, or Team features.

Short version: pick the model that matches your immediate question, then connect it in Workflows.

• For fast protein folding, start with Chai-1.
• To rank or score complexes, use DockQ.
• For sequence design on an existing complex, use AntiBMPNN.
• For generative sequence exploration, explore Boltz-2 and BoltzGen.
• To compare against strong structure baselines, run OpenFold-3 or RoseTTAFold-3.
• For surface and binding patch context, add ParaSurf to your workflow.

For the full and always updated list of models, visit the platform library.

Yes. You can start from dedicated use case pages for proteins, ligands, and nucleic acids.

Each page highlights recommended models, example workflows, and templates that match common tasks in that space.

Yes. Every time you set up a run in the interface you can click Sync to see the matching API call. Sync turns your chosen inputs and settings into ready to use code that you can copy into your own tools or notebooks.

Your personal API key is available on your account page. The same credit system applies, so jobs started by the API use credits in exactly the same way as jobs started in the browser.

If you prefer to browse models first and then script them, you can explore everything in the platform library, then use Sync to generate API code when you are ready.

Kanban is your project board inside Vici.bio. You can group jobs into columns such as Ideas, In progress, Review, Done, see what is running, and keep the lab or team on the same page.

Kanban is available to all plans. It is designed so that biologists can manage work visually without needing to use external project tools.

AI Agent is a chat assistant that knows your projects, jobs, and models inside Vici.bio. It answers questions using information from our private systems, not from the public internet.

The agent is available to all users and uses the same credit system. Each question uses credits and is processed one at a time, so you keep control of usage.

In future versions, the agent will also be able to draft and configure workflows for you, so you can go from a plain language question to a complete pipeline.

Editor is a simple sequence editor built for biologists. You can view sequences, select regions, and adjust constructs before sending them to models or workflows.