Flower Agent

What it does

Flower Agent provides a runtime for agentic applications built as AgentApps. Developers supply agent logic while Flower provides model access, connectors and run state. Applications can run locally or on SuperGrid, with browser and terminal chat interfaces for interaction.

Connectors expose runtime tools and authorized account access to the agent. Related runs can form a conversation, and developers can publish applications through Flower Hub. The product is experimental, so interfaces may change; its internal OpenAI-compatible runtime endpoint is not a public inference API for arbitrary clients.

What it can do

• AgentApp runtime
• Browser chat
• Terminal chat
• Account connectors
• Run-state management
• Local execution

Before you choose

• Experimental interfaces and behavior may change between releases.
• Confirm hosted compute and model-use terms for the chosen environment.

Listed under

• Coding
• MCP and agent tools

Get started

1. Choose browser chat, terminal chat or a custom AgentApp tutorial.
2. Configure the runtime and authorize only the connectors needed.
3. Test the application and review its behavior before operational use.

The documentation describes local and SuperGrid execution. Confirm model and compute allowances for the selected deployment; no numeric price is established here.

We couldn’t find published prices for Flower Agent. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 30 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Flower Agent ↗flower.ai

Company website

flower.ai: product, pricing and policy pages

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

manual_review

Last checked 30 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.