flywheel-ai/automotive

TEXT GENERATIONConcurrent Unit Cost:3Model Size:35.1BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 21, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The flywheel-ai/automotive model is a 35 billion parameter vertical AI-employee model developed by Flywheel by OpSpot. Fine-tuned from Qwen/Qwen3.6-35B-A3B, it specializes in the automotive domain, acting as a service advisor assistant for auto repair shops and dealerships. Its primary function is to assist with diagnostic intake, service scheduling, estimates, parts lookup, and customer updates, making it highly optimized for specific automotive service tasks.

Loading preview...

Overview

flywheel-ai/automotive is an open-source vertical AI-employee model developed by Flywheel by OpSpot. It is fine-tuned (LoRA) from the Qwen/Qwen3.6-35B-A3B base model, specifically designed for the automotive domain.

Key Capabilities

This model functions as a practical service-advisor assistant for auto repair shops and dealership service departments, offering support for:

  • Diagnostic intake: Assisting with initial vehicle problem assessment.
  • Service scheduling: Managing appointments and service timelines.
  • Estimates and pricing: Generating cost estimates for repairs and services.
  • Parts lookup: Identifying and sourcing necessary automotive parts.
  • Maintenance schedules: Providing information on routine vehicle maintenance.
  • Recall awareness: Keeping track of vehicle recalls.
  • Customer status updates: Communicating repair progress to customers.

Technical Details & Provenance

The model is based on Qwen3.6-35B-A3B and is available in safetensors format (for transformers/vLLM, approximately 65GB) and model-q4_k_m.gguf (for llama.cpp/Ollama, approximately 20GB). The initial v1.0 training utilized synthetic seed data generated by permissively-licensed local models (Apache/MIT teachers only), ensuring no distillation from closed models. While its general performance is comparable to the base model, its specialized automotive niche is expected to improve with real-world usage.

Guardrails

It includes a guardrail that defers safety-critical systems (e.g., brakes, steering, airbags) to certified mechanics and issues do-not-drive warnings when a vehicle may be unsafe, emphasizing that it is not a certified technician.

Finetunes
1 models