You can run ~1b-7b 4-bit models locally on an iPhone with MLX Swift.
https://github.com/ml-explore/mlx-swift-examples/tree/main/Applications/LLMEval
You can run ~1b-7b 4-bit models locally on an iPhone with MLX Swift.
https://github.com/ml-explore/mlx-swift-examples/tree/main/Applications/LLMEval
Good to know! Thanks