flywheel-ai/real-estate

Flywheel — real-estate (35b-v1)

An open-source vertical AI-employee model from Flywheel by OpSpot, fine-tuned (LoRA) from Qwen/Qwen3.6-35B-A3B (Apache-2.0) for the real-estate domain.

Practical operations assistant for licensed real-estate agents: listing intake and descriptions, lead capture and follow-up, showing scheduling, transaction-coordination checklists, market-info FAQs, and vendor referrals.

• —Base: Qwen/Qwen3.6-35B-A3B · License: Apache-2.0 · Version: 35b-v1
• —Formats: safetensors (transformers / vLLM, ~65G) + model-q4_k_m.gguf (llama.cpp / Ollama, ~20G)

Download (one command)

pip install -U huggingface_hub
hf download flywheel-ai/real-estate                      # full repo (safetensors + GGUF)
hf download flywheel-ai/real-estate model-q4_k_m.gguf    # just the GGUF

Run

# llama.cpp
llama-server -m model-q4_k_m.gguf -ngl 999
# Ollama (pulls the GGUF straight from HF)
ollama run hf.co/flywheel-ai/real-estate
# vLLM (serves the safetensors)
vllm serve flywheel-ai/real-estate

Guardrail

Complies with the Fair Housing Act (no steering or protected-class preference language); legal, financing, and valuation specifics defer to a licensed attorney, lender, or appraiser.

Provenance & honesty

v1.0 is trained on synthetic seed data authored by permissively-licensed local models (Apache/MIT teachers only — never distilled from closed models). On general prompts it is roughly on par with the base; the niche edge sharpens as consented real usage flows through the OpSpot flywheel. Built on Qwen3.6 (Apache-2.0).