Model HQ

Documentation

Apple Supported Models

Comprehensive list of 120+ AI models optimized for Apple Silicon and Apple Neural Engine

These models are optimized for Apple Silicon processors and leverage Metal Performance Shaders and the Apple Neural Engine for enhanced performance on macOS and iOS devices.

Apple Optimization Features

Key benefits of Apple-optimized models

Performance Benefits

• Optimized for Apple Silicon architectures (M-series chips)
• Enhanced inference speed with Metal Performance Shaders
• Power-efficient execution on Apple Neural Engine
• Unified memory architecture for fast data access

Supported Hardware

• Apple Silicon M1, M2, M3, and M4 series processors
• Apple Neural Engine (ANE)
• Apple integrated GPUs
• macOS and iOS devices

Apple Supported Models

Complete catalog of models optimized for Apple processors

Model Type	Available Models
Embedding Models	`all-mini-lm-L6-v2` `all-mpnet-base-v2` `industry-bert-insurance` `industry-bert-contracts` `industry-bert-asset-management` `industry-bert-sec` `industry-bert-loans` `nomic-ai/nomic-embed-text-v1`
Jina Models	`jina-reranker-tiny-ppt` `jina-reranker-turbo-ppt` `jina-reranker-tiny-onnx` `jina-reranker-turbo-onnx` `jina-reranker-v1-turbo-en` `jina-reranker-v1-tiny-en`
GPT Models	`gpt-5.2-pro` `gpt-5.2` `gpt-5-mini` `gpt-5-nano` `gpt-4.1`
Claude Models	`claude-opus-4-5` `claude-haiku-4-5` `claude-sonnet-4-5` `claude-sonnet-4-20250514` `claude-opus-4-20250514`
Gemini Models	`gemini-3-pro-preview` `gemini-3-flash-preview` `gemini-2.5-pro` `gemini-2.5-flash` `gemini-2.5-flash-lite`
Qwen Models	`qwen2.5-vl-3b-instruct-gguf` `qwen3-vl-8b-gguf` `qwen3-vl-4b-gguf` `qwen3-vl-30b-gguf` `bling-qwen-mini-tool` `bling-qwen-0.5b-gguf` `dragon-qwen-7b-gguf` `qwen2-7B-instruct-gguf` `qwen3-1.7b-gguf` `qwen3-4b-instruct-gguf` `qwen3-8b-gguf` `qwen3-14b-gguf` `qwen-3.5-4b-gguf` `qwen-3.5-9b-gguf` `qwen-3.5-27b-gguf` `qwen-3.5-35b-a3b-gguf` `qwen2-1.5b-instruct-gguf` `qwen2-0.5b-instruct-gguf` `qwen-2.5-7b-coder-gguf` `qwen-2.5-14b-instruct-gguf` `qwen2.5-32b-gguf` `qwen2.5-72b-gguf` `deepseek-qwen-14b-gguf` `deepseek-qwen-7b-gguf`
Llama-Based Models	`llama-3.1-instruct-gguf` `llama-2-7b-chat-gguf` `llama-3-8b-instruct-gguf` `tiny-llama-chat-gguf` `llama-3.2-1b-instruct-gguf` `llama-3.2-3b-instruct-gguf` `dragon-llama-3.1-gguf`
Phi Models	`bling-phi-3-gguf` `bling-phi-3.5-gguf` `phi-3.5-gguf` `phi-4-gguf` `phi-4-mini-gguf` `phi-4-mini-reasoning-gguf` `phi-3-gguf` `phi-3-ppt`
Mistral Models	`dragon-mistral-0.3-gguf` `mistral-small-3.2-24b-gguf` `ministral-3-14b-gguf` `openhermes-2.5-mistral-7b-gguf` `zephyr-7b-beta-gguf` `starling-lm-7b-alpha-gguf` `mistral-7b-instruct-v0.3-gguf`
Yi Models	`dragon-yi-9b-gguf` `dragon-yi-answer-tool`
Gemma Models	`gemma-3-4b-gguf` `gemma-3-12b-gguf` `gemma-4-4b-gguf` `gemma-4-2b-gguf` `gemma-4-26b-gguf` `gemma-2-9b-instruct-gguf` `gemma-2-27b-instruct-gguf`
StableLM Models	`bling-stablelm-3b-gguf`
Dragon Models	`dragon-llama-3.1-gguf` `dragon-mistral-0.3-gguf` `dragon-yi-9b-gguf` `dragon-qwen-7b-gguf` `bling-qwen-mini-tool` `bling-qwen-0.5b-gguf` `dragon-yi-answer-tool` `dragon-llama-answer-tool` `dragon-mistral-answer-tool`
Slim Models	`slim-ner-tool` `slim-sentiment-tool` `slim-emotions-tool` `slim-ratings-tool` `slim-intent-tool` `slim-nli-tool` `slim-topics-tool` `slim-tags-tool` `slim-sql-tool` `bling-answer-tool` `slim-category-tool` `slim-xsum-tool` `slim-extract-tool` `slim-extract-phi-3-gguf` `slim-extract-qwen-1.5b-gguf` `slim-extract-qwen-nano-gguf` `slim-extract-tiny-tool` `slim-summary-tiny-tool` `slim-summary-phi-3-gguf` `slim-xsum-phi-3-gguf` `slim-boolean-tool` `slim-boolean-phi-3-gguf` `slim-sa-ner-phi-3-gguf` `slim-sa-ner-tool` `slim-tags-3b-tool` `slim-summary-tool` `slim-q-gen-phi-3-tool` `slim-q-gen-tiny-tool` `slim-qa-gen-tiny-tool` `slim-qa-gen-phi-3-tool`
Specialized Models	`gpt-oss-20b-gguf` `olmo-13b-gguf` `granite-4-micro-gguf` `liquidai-lfm2-2.6b-gguf` `minicpm-2.6-gguf` `whisper-cpp-base-english`

Getting Started with Apple Models

To use Apple-optimized models in Model HQ:

Ensure you have a device with Apple Silicon (M1 or later)
Select models optimized for Apple from the Models section
The system will automatically use Apple optimizations when available
Monitor performance improvements and power efficiency

Technical Support

For Apple-specific optimization questions or issues, contact our technical support team at support@aibloks.com

🚀 Performance Tip

Apple-optimized models are designed for efficient on-device computing. They provide excellent performance while maintaining low power consumption thanks to Apple Silicon's unified memory architecture and Neural Engine.

Check System Requirements