swiftlet
Swiftlet runs large Qwen MoE models locally on Apple devices with a Swift and Metal runtime, reducing RAM use for on-device inference.
Share on XLicense: Apache-2.0
Overview
Swiftlet is a Swift and Metal runtime for Qwen3-Next and Qwen3.5/3.6 Mixture-of-Experts models. It keeps the dense core in memory and streams routed expert weights from storage on demand, so large models can run on ordinary Apple devices while using less RAM. It is used for local inference on macOS and iOS, including supported iPhone setups, with a focus on practical local execution rather than cloud-based serving.
Key features
- Runs large Qwen MoE models locally on macOS and iOS devices
- Keeps the dense model core in memory and streams expert weights from storage on demand
- Targets 35B and 80B model sizes in the project examples
- Uses a Swift and Metal runtime for on-device inference
Best for
Best for developers and hobbyists working with local AI inference on Apple devices, especially when a model must fit within tight RAM limits. It is a practical option for macOS or iOS setups that need to run large Qwen MoE models without relying on cloud hosting.
- Upstream
- leonickson1/Swiftlet
- Fork on GitHub
- Guo-astro/swiftlet
- Upstream stars
- 654
- Category
- AI agents and LLM tools
- License
- Apache-2.0
- Forked
- 2026-10-05
- Sync status
- In syncLast synced 2026-10-11
More in AI agents and LLM tools
Hello-Agents is a tutorial for learning the principles and practice of AI agents. It guides readers from core concepts to building agent systems and applications.
Forked 2026-10-10Last synced 2026-10-11Custom license (see repository)AI agents and LLM toolsGitHub
OpenSRE is an open-source framework for building AI agents that help investigate production incidents using your infrastructure tools and workflows.
Forked 2026-10-10Last synced 2026-10-11License: Apache-2.0AI agents and LLM toolsGitHub
Context Mode helps AI coding agents use less context and resume work after a conversation is compacted. It is used through MCP and hooks to handle tool output, session memory, and routing across supported platforms.
Forked 2026-10-10Last synced 2026-10-11Custom license (see repository)AI agents and LLM toolsGitHub
ARTEX is a self-hosted AI penetration testing system for exploring assets and reviewing test activity. It brings tasks, findings, asset data and approval steps into one interface.
Forked 2026-10-10Last synced 2026-10-11License: AGPL-3.0AI agents and LLM toolsGitHub