Alex Cheema
Alex Cheema
## Motivation Currently the macOS app requires internet to start, since the sparkle update runs on start in the foreground. ## Changes Run sparkle updater in background. ## Why It...
Currently the macOS app runs some swift code / a bash script on start to bootstrap RDMA but this is not documented when running from source. There are known issues...
## Motivation Often users are running into issues with RDMA. See https://github.com/exo-explore/exo/issues?q=is%3Aissue%20rdma Having some debug info in the macOS app will help to debug these issues. ## Changes Displays output...
Other apps like Ollama, LMStudio have their own model directory. exo should be able to load models from their directories. Related to https://github.com/exo-explore/exo/issues/1011
## Motivation Discord link expired. ## Changes Replace discord invite link with permanent link. ## Why It Works It's permanent now. ## Test Plan Clicked the link. It works.
## Describe the bug When launching a lot of instances in succession, we run into `Too many open files` errors. ## To Reproduce Steps to reproduce the behavior: 1. Launch...
## Describe the bug Launching an instance of `mlx-community/Qwen3-30B-A3B-4bit` on 2 nodes with Tensor/RDMA gets stuck in LOADING. ## To Reproduce Steps to reproduce the behavior: 1. Start exo on...
## Motivation Upgrade mlx-lm to version 0.30.2 which requires transformers 5.0.0rc2 as a prerelease dependency. This enables support for newer models like Kimi K2 Thinking while maintaining compatibility with existing...
## Motivation Claude Code (claude.ai/code) uses CLAUDE.md files to understand project context. This file helps future Claude Code sessions be more productive by providing build commands, architecture overview, and code...
## Motivation Add support for the GPT-OSS model architecture in the MLX engine. This model requires a custom chat template that isn't included in the model repository, so we bundle...