Gemma 3 is a lightweight open-source multimodal model launched by Google. This version is an instruction-tuned quantization-aware training model with 12B parameters, which has been converted to the MXFP4 format of the MLX framework. It supports text and image input and generates text output, with a 128K context window and support for over 140 languages.

MultimodalMlx

This is an automatic speech recognition model optimized for Apple silicon chip devices. By converting to the MLX framework and quantizing to the FP8 format, it enables fast on-device speech transcription on Apple devices. The model is fine-tuned for verbatim accuracy and is particularly suitable for scenarios requiring high-precision transcription.

Audio ProcessingMlxOther

This model is an 8-bit quantized version converted from allenai/Olmo-3-7B-Instruct, specifically optimized for the Apple MLX framework. It is a large language model with 7 billion parameters, supporting instruction following and dialogue tasks.

Natural Language ProcessingMlxEnglish

The 4-bit quantized version of VibeThinker-1.5B, optimized for Apple chips based on the MLX framework, is a dense language model with 1.5 billion parameters, specifically designed for mathematical reasoning and algorithm coding problems.

Natural Language ProcessingMlxMultiple Languages

This model is an MLX format model converted from summykai/gemma3-27b-abliterated-dpo using version 0.28.3 of mlx-lm. It is a 27B parameter Gemma 3 large language model fine-tuned with DPO (Direct Preference Optimization), optimized for efficient operation on Apple Silicon (MLX framework).

Natural Language ProcessingTransformersEnglish

This model is an MLX format conversion version of the instruction fine-tuned version of Falcon-H1-34B-Instruct, optimized specifically for Apple Silicon (M series chips). It is based on the original Falcon-H1-34B-Instruct model and converted to an 8-bit quantization format compatible with the MLX framework through the mlx-lm tool, aiming to achieve efficient local inference on macOS devices.

Natural Language ProcessingTransformersMultiple Languages

This is a text-to-speech model optimized based on the MLX framework, converted from the original model Marvis-AI/marvis-tts-100m-v0.2. It uses 6-bit quantization technology and is specifically optimized for Apple Silicon hardware, providing efficient speech synthesis capabilities.

Audio ProcessingTransformersMultiple Languages

Qwen3-Coder-480B-A35B-Instruct is a large code generation model with 480 billion parameters, supporting 8.5-bit quantization and optimized based on the MLX framework. This model is specifically designed for code generation tasks and can run efficiently on devices with sufficient memory.

Natural Language ProcessingMlxEnglish

This is a 6-bit quantized version converted from the Kimi-Linear-48B-A3B-Instruct model, optimized for the Apple MLX framework. The model retains the powerful instruction-following ability of the original model, while significantly reducing storage and computational requirements through quantization technology, making it suitable for efficient operation on Apple hardware.

Natural Language ProcessingMlx

This model is based on the Kimi-Linear-48B-A3B-Instruct large language model of moonshotai and is converted into an 8-bit quantized version suitable for Apple Silicon chips (MLX framework) using the mlx-lm tool. It is an instruction fine-tuned model with 48B parameters, designed specifically for following human instructions and dialogue interaction.

Natural Language ProcessingMlx

This is a 4-bit quantized version converted from the moonshotai/Kimi-Linear-48B-A3B-Instruct model, optimized for the Apple MLX framework, providing efficient text generation capabilities.

Natural Language ProcessingMlx

MiniMax-M2-5bit is a 5-bit quantized version converted from the MiniMaxAI/MiniMax-M2 model, optimized specifically for the MLX framework. This model is a large language model that supports text generation tasks and is released under the MIT license.

Natural Language ProcessingMlx

This is a high-performance 4-bit quantized version of the Kimi K2 Instruct model, optimized for running on Apple Silicon (M1/M2/M3/M4) Macs using the MLX framework. The model has 671 billion parameters, supports a 128K context window, and achieves an excellent balance between quality and efficiency, making it an ideal choice for most practical deployments.

Natural Language ProcessingMlxMultiple Languages

This is a 4-bit quantized version converted from the Qwen3-VL-32B-Thinking model, specifically optimized for the MLX framework. This model is a multimodal vision-language model with a parameter scale of 32B. It has the ability of chain-of-thought reasoning and can process both image and text inputs simultaneously to generate high-quality text responses.

MultimodalTransformers

This is the 4-bit quantized version of the Qwen3-VL-4B-Instruct model, specifically optimized for Apple Silicon chips and converted using the MLX framework. This model is a vision-language model that supports image understanding and multimodal dialogue tasks.

MultimodalMlx

This is an 8-bit quantized version converted from the LiquidAI/LFM2-8B-A1B model, optimized for the Apple MLX framework. This model is a mixture-of-experts model with 8B parameters, supporting multilingual text generation tasks.

Natural Language ProcessingMlxMultiple Languages

ethicalabs/granite-4.0-h-small-base-MLX is an MLX format version converted from the IBM Granite-4.0-H-Small-Base model, specifically optimized for the Apple MLX framework to provide efficient text generation capabilities.

Natural Language ProcessingMlx

This is a text-to-speech model optimized based on the MLX framework, developed by nineninesix, supporting high-quality speech synthesis in multiple languages.

Audio ProcessingMlxMultiple Languages

This is a 4-bit quantized version of the IBM Granite-4.0-h-Tiny model, optimized for Apple Silicon and using the MLX framework for efficient inference. The model has been processed with DWQ (Dynamic Weight Quantization) to significantly reduce the model size while maintaining performance.

Natural Language ProcessingMlx

This is a 4-bit quantized version converted from the Apriel-1.5-15b-Thinker model based on ServiceNow-AI, specifically optimized for the MLX framework and supporting visual language understanding tasks.

MultimodalTransformers