flywheel-ai/fitness
The flywheel-ai/fitness model is a 35.1 billion parameter vertical AI-employee model developed by Flywheel by OpSpot. Fine-tuned from Qwen/Qwen3.6-35B-A3B, it specializes as a knowledgeable and encouraging fitness coach. This model excels in providing guidance on strength and hypertrophy programming, fat loss, nutrition, mobility, recovery, conditioning, and habit change within the fitness domain.
Loading preview...
Overview
Flywheel-ai/fitness is a 35.1 billion parameter model developed by Flywheel by OpSpot, specifically fine-tuned for the fitness domain. It is based on the Qwen/Qwen3.6-35B-A3B architecture and is designed to act as a knowledgeable and encouraging fitness coach. The model is available in safetensors format for transformers/vLLM and GGUF format for llama.cpp/Ollama.
Key Capabilities
- Fitness Coaching: Provides guidance on strength and hypertrophy programming, fat loss, nutrition, mobility, recovery, conditioning, and habit change.
- Open-Source Provenance: v1.0 is trained on synthetic seed data authored by permissively-licensed local models (Apache/MIT teachers only), ensuring no distillation from closed models.
- Guardrails: Includes a guardrail stating that it does not provide medical advice and defers to licensed professionals for pain, injury, illness, medication, or pregnancy.
Use Cases
- Personalized Fitness Guidance: Ideal for applications requiring an AI to offer detailed and encouraging advice on various aspects of physical fitness and well-being.
- Fitness App Integration: Can be integrated into fitness applications to provide users with AI-driven coaching and support.
- Research and Development: Suitable for developers and researchers working on specialized AI models for health and wellness, leveraging its open-source training methodology.