alexgusevski/saiga_yandexgpt_8b-mlx

TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kPublished:Jan 12, 2026License:yandexgpt-5-lite-8b-pretrainArchitecture:Transformer Cold

The alexgusevski/saiga_yandexgpt_8b-mlx is an 8 billion parameter language model, converted to the MLX format by alexgusevski from the original IlyaGusev/saiga_yandexgpt_8b. This model is specifically designed for efficient deployment and inference on Apple Silicon, leveraging the MLX framework. Its primary utility lies in providing a readily available, optimized version of the Saiga YandexGPT 8B model for MLX-compatible environments.

Loading preview...

Overview

The alexgusevski/saiga_yandexgpt_8b-mlx model is a specialized conversion of the IlyaGusev/saiga_yandexgpt_8b language model, optimized for use with Apple Silicon via the MLX framework. This 8 billion parameter model facilitates local, efficient inference on compatible hardware.

Key Characteristics

  • MLX Format: Converted using mlx-lm version 0.30.2, ensuring compatibility and performance on Apple Silicon.
  • Base Model: Derived from the IlyaGusev/saiga_yandexgpt_8b model, indicating its foundational capabilities.
  • Ease of Use: Provides straightforward integration with the mlx-lm library for loading and generating text.

Use Cases

  • Local Inference: Ideal for developers and researchers looking to run the Saiga YandexGPT 8B model directly on their Apple Silicon devices.
  • MLX Development: Serves as a practical example and resource for projects utilizing the MLX ecosystem for language model deployment.
  • Experimentation: Suitable for experimenting with the Saiga YandexGPT 8B model's capabilities in an optimized local environment.

Finetunes
1 models