---
title: Get started
description: >-
  Start with guest chat or curl, then add an account when you need saved
  sessions, higher quota, and the tools your plan allows.
tags:
  - getting-started
  - onboarding
  - subscription-tiers
  - workspace
---

Chat is one way to start a run; curl, the CLI and an MCP client reach the same tools. This guide shows how to try each one without an account, what an account adds, and where Studio fits.

> **In a hurry?**

> Start a [guest chat](https://www.ivybiosciences.com/chat/new?guest=true), or run the [guest curl example](/docs/api#guest-curl). Sign up at [ivybiosciences.com/signup](https://www.ivybiosciences.com/signup) when you need persistence, then jump to the [ESMFold quickstart](/docs/esmfold).

## What an account adds

Guest Chat and curl are useful for trying the platform, but the account path is where work becomes a durable project:

Without an account, you can still submit guest Chat and curl requests within the public limits. Guest work is temporary and does not provide the saved sessions, project history, or authenticated API/MCP access that an account provides.

## Create your account

    **Sign up.** Go to [ivybiosciences.com/signup](https://www.ivybiosciences.com/signup) and either create an email + password, or sign in with Google/GitHub OAuth.

    **Verify your email.** Click the confirmation link in your inbox. Check spam if it hasn't arrived in 5 minutes.

    **Complete your profile.** Add your name and organization (optional). Students can choose the [student plan](/pricing#student).

    **Start on your plan.** New accounts begin with the current entry tier and its credit/tool limits; see [Plans and tiers](/docs/plans-and-tiers) rather than relying on an older job-count label.

    **Run something.** Submit through [Chat](/docs/chat), use the [API](/docs/api), or install the [CLI](/docs/cli) and run `ivybloom run esmfold protein_sequence=MKFLILLFNILCLFPVLAADNH --follow`.

## Subscription tiers

Plans meter work with credits and gate model families, compute tools, wetlab tools, and team features. API access is not gated: every account, including the free Seed plan, can create an API key. The [Plans and tiers](/docs/plans-and-tiers) page is the canonical summary, and the pricing page remains authoritative for current commercial rates.

> **Guest before signup**

> Guest Chat and curl start without a card or account. Sign in when you need saved sessions, higher quota, the tools your plan allows, or the authenticated [API and MCP surfaces](/docs/api).

## Web and terminal workspaces

Chat and Studio are the browser workspaces; the CLI is the terminal surface. All three connect to the same registry, and authenticated sessions share job history. Curl and MCP are the programmatic paths described in the [entry-point overview](/docs/overview).

**Chat / Studio**

The browser workspace at [ivybiosciences.com/chat](https://www.ivybiosciences.com/chat).

**Best for**
- Interactive exploration and visualization
- Running individual analyses
- 3D molecular structures (Mol* integration)
- Sharing results with non-technical collaborators
- Quick one-off jobs

**Key features**
- Point-and-click tool selection
- Real-time job status with progress streaming
- Interactive 3D viewers (Mol*, 3DMol, NGL)
- Result comparison dashboards
- Multi-format export (PDB, SDF, CSV, PNG, JSON)

**Getting started**

  Sign in at [ivybiosciences.com](https://www.ivybiosciences.com/login).
  Pick a tool from the sidebar (e.g. ESMFold, DiffDock, REINVENT4).
  Upload your input or paste a sequence / SMILES / PDB.
  Configure parameters and click **Submit**.
  Inspect results as a canvas card when the job completes.

**CLI**

The [ivybloom CLI](/docs/cli) for terminal workflows.

**Best for**
- Batch processing many inputs
- Scripting and automation
- CI/CD pipeline integration
- Programmatic submission with custom orchestration
- Power users who prefer the terminal

**Key features**
- Submit jobs from shell scripts
- Watch job status live (`ivybloom jobs watch`)
- Auto-download completed artifacts
- Pipe outputs to other tools (`jq`, `xargs`, etc.)
- JSON / CSV output for scripting

**Getting started**

  Install the CLI — see [Install the CLI](#install-the-cli) below.
  Authenticate: `ivybloom auth login --browser`
  List available tools: `ivybloom tools list`
  Submit a job: `ivybloom run esmfold protein_sequence=MKFLILLFNILCLFPVLAADNH --follow`
  Check status: `ivybloom jobs list` or `ivybloom jobs status <job_id>`

### Install the CLI

**npm**

```bash
npm install -g @ivybiosciences/ivybloom-cli
ivybloom auth login --browser
```

**pnpm**

```bash
pnpm add -g @ivybiosciences/ivybloom-cli
ivybloom auth login --browser
```

**bun**

```bash
bun add -g @ivybiosciences/ivybloom-cli
ivybloom auth login --browser
```

### Which workspace should I use?

> **Pro tip**

> Many researchers use both. Start in Chat or Studio to understand a tool's inputs/outputs and visualize the first result, then switch to the CLI for production batch workflows.

## How jobs execute

Whether you submit from Chat, Studio, or the CLI, every job flows through the same pipeline:

 Q["Queue"] --> P["GPU Processing"] --> R["Results Ready"]
    S --> V["Validation"]
    Q --> PR["Priority<br/>(by tier)"]
    P --> AI["AI Model<br/>Execution"]
    R --> NT["Notification<br/>+ Download"]`}
  caption="Every job passes through submission, queue, GPU processing and results."
/>

### Job states

### Viewing job history

**Chat / Studio**

    Navigate to **Jobs** in the sidebar. Each row links to the canvas card view; filters for status, tool, and date range are available.

**CLI**

```bash
ivybloom jobs list                    # all your jobs, paginated
ivybloom jobs list --status running   # active only
ivybloom jobs results <job_id>        # results and artifacts of a completed job
ivybloom jobs cancel <job_id>         # cancel in-flight
```

Result retention and credit limits vary by plan. See [Plans and tiers](/docs/plans-and-tiers) for the current source of truth.

## Common questions

### Do I get charged per job on the Free tier?

    Guest runs and authenticated plans use different limits. Read the current credit and access rules in [Plans and tiers](/docs/plans-and-tiers); do not treat an older job-count label as the source of truth.

### Can I upgrade or downgrade mid-month?

    Yes. Upgrades take effect immediately and bill prorated. Downgrades take effect at the end of the current billing cycle; your existing quota remains active until then.

### What happens to my data if I cancel?

    Job results remain accessible (read-only) for the retention window of your final tier. After that they're permanently deleted. Account metadata (email, settings) is retained for 90 days then purged unless you re-subscribe.

### Is my data used to train models?

    No. Submitted inputs and produced artifacts are private to your account. Enterprise customers can sign a zero-data-retention addendum that purges inputs from logs and ephemeral caches immediately on completion.

### Can I run Ivy Biosciences offline / on-prem?

    Self-hosting is available through an enterprise arrangement; ask [support](/help/support). The platform is licensed FSL-1.1-Apache-2.0 (each release converts to Apache-2.0 two years after it ships), and a public source release is planned; the repositories are not public yet.

## Read next

## Getting help

- **Documentation** — browse the sidebar for tool-specific guides, or press <kbd>⌘K</kbd> to search.
- **Troubleshooting** — [common issues and solutions](/docs/troubleshooting).
- **Support** — open a ticket from [support](/help/support).