Meta Preps Paid AI Agent 'Hatch' for Launch, Plans New Model 'Watermelon' for October
Meta is preparing to launch Hatch, a consumer AI agent that could cost up to $199.99/month, alongside a new model called Watermelon slated for October. Details come from internal documents cited by The Information, with key specifications still unconfirmed.
Meta Platforms is preparing to launch a paid consumer AI agent called Hatch in the coming weeks, alongside a new AI model named Watermelon expected in October, according to internal documents reviewed by The Information.
Hatch is described as a consumer-facing version of Meta's internal OpenClaw agent project. The move is part of CEO Mark Zuckerberg's push to generate direct revenue from Meta's AI investments rather than relying solely on advertising income. According to the report, Meta is considering a tiered subscription structure, with a premium tier potentially priced as high as $199.99 per month. Pricing for lower tiers has not been disclosed.
Hatch is designed to connect with third-party services including DoorDash, Etsy, Reddit, Yelp, and Outlook. The agent will reportedly feature a customizable dashboard supporting modules such as a fitness tracker and a trip planner, positioning it as a general-purpose personal assistant rather than a narrow chatbot.
Separately, Meta plans to release a new model called Watermelon in October. It remains unclear, according to the report, whether Watermelon will be folded into Meta's existing Muse model family or released as a standalone model. No technical specifications — including parameter count, context window, training cutoff, or benchmark scores — have been disclosed. Meta has not publicly confirmed the model's existence or capabilities.
Meta is also reportedly building a platform within WhatsApp that would allow third-party AI agents to plug into the messaging app, expanding the potential distribution surface for Hatch and other agents beyond Meta's own apps.
The information in this report originates from internal Meta documents obtained by The Information, not from an official Meta announcement. Meta has not confirmed pricing, launch timing, or technical details for either Hatch or Watermelon on the record.
What this means
This is a leak-based report, not a product launch — key facts like final pricing, Watermelon's architecture, and exact release dates remain unconfirmed and could change before anything ships. The $199.99/month figure, if accurate, would place Hatch well above existing AI subscription tiers from OpenAI and Anthropic, signaling Meta may be targeting power users or enterprise-adjacent consumers rather than mass-market adoption. The bigger signal is strategic: Meta is trying to build a monetizable agent layer on top of its LLMs, tying model capability directly to consumer-facing products that access real-world services like DoorDash and Yelp. Whether Watermelon is a new foundation model or a variant of Muse will matter significantly for how Meta's model lineup is understood — a standalone release would suggest a broader architecture shift, while a Muse variant would suggest incremental scaling of an existing line.
Related Articles
AWS Publishes Reference Architecture for Multimodal WhatsApp Ordering Agents Using Bedrock AgentCore and Nova 2
AWS published a reference architecture showing how to deploy a WhatsApp ordering assistant on Amazon Bedrock AgentCore, using Nova 2 Lite for text and Nova 2 Sonic for voice, with shared cross-channel memory and MCP-based tool access to backend systems.
Google Launches Pics, a Prompt-Based Design Tool Built Into Workspace
Google is launching Pics, an AI image creation and editing tool powered by its Nano Banana model, positioning it as a prompt-first alternative to Canva and Adobe Express. The tool rolls out first in Google Docs and Slides for Workspace and Google AI Pro/Ultra subscribers, with Drive support to follow.
Gemini Overlay on Android Adds Minimize Button for Multitasking Bubble
Google is widely rolling out a new Minimize button for the Gemini overlay on Android, which collapses conversations into a floating bubble users can drag or tap to expand. The feature currently supports six fixed positions and has a reported bug that resets bubble placement after each minimization.
OpenAI Lists GPT-6 Astra Pro on OpenRouter: Same Model, Higher-Compute Reasoning Mode
GPT-6 Astra Pro, now listed on OpenRouter, is the existing GPT-6 Astra model configured to run with reasoning.mode set to 'pro' for higher-quality output on complex tasks. It carries a 1M-token context window and tiered pricing from $5/$25 to $20/$100 per million input/output tokens depending on the serving tier.
Comments
Loading...