flywheel-ai/real-estate

TEXT GENERATIONPricing:Input $0.4 / Cached $0.07 / Output $4Concurrent Unit Cost:3Model Size:35.1BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 23, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The flywheel-ai/real-estate model, developed by Flywheel by OpSpot, is a 35.1 billion parameter language model fine-tuned from Qwen/Qwen3.6-35B-A3B with a 32768 token context length. It is specifically optimized as a vertical AI assistant for the real estate domain, designed to assist licensed real estate agents with tasks like listing management, lead follow-up, and scheduling. This model excels in practical operations support within the real estate industry, adhering to Fair Housing Act guidelines.

Loading preview...

Overview

Flywheel by OpSpot's flywheel-ai/real-estate is a specialized 35.1 billion parameter AI model, fine-tuned from the Qwen/Qwen3.6-35B-A3B base model. It functions as a vertical AI-employee specifically designed for the real estate sector, offering practical assistance to licensed agents. The model has a substantial context length of 32768 tokens, enabling it to handle complex and detailed real estate-related queries and tasks.

Key Capabilities

  • Listing Management: Assists with intake and description generation for property listings.
  • Lead Management: Supports lead capture and follow-up processes.
  • Scheduling: Facilitates showing scheduling.
  • Transaction Coordination: Provides checklists for transaction coordination.
  • Information Retrieval: Answers FAQs related to market information.
  • Vendor Referrals: Offers vendor referral suggestions.
  • Compliance: Incorporates guardrails to comply with the Fair Housing Act, avoiding steering or protected-class preference language.

Provenance and Training

The model's v1.0 was trained using synthetic seed data authored by permissively-licensed local models (Apache/MIT licensed teachers only, not distilled from closed models). It is built upon the Apache-2.0 licensed Qwen3.6 architecture. While its general performance is comparable to its base model, its niche capabilities are expected to sharpen with consented real-world usage through the OpSpot flywheel.

When to Use This Model

This model is ideal for real estate professionals seeking an AI assistant to streamline daily operations. It is particularly well-suited for tasks requiring domain-specific knowledge in real estate, such as drafting property descriptions, managing client interactions, and ensuring compliance with housing regulations. Users should note that for legal, financing, and valuation specifics, the model defers to licensed professionals.

Finetunes
1 models