An edge-native MoE serving engine that runs large open-weight models like DeepSeek-V4 and Qwen3.6 on consumer GPUs by splitting work between GPU and CPU memory.
14.6K1.5KPythonApache-2.0Updated 16 hours ago
2 alternatives
Looking for a free, open-source alternative to LM Studio? These 2 projects can replace it. Each one is reviewed by a human and explained in plain words, so you can compare stars, license and maintenance at a glance.
| Project | Stars | License | Last commit |
|---|---|---|---|
| FlashML-org/FreeToken | ★ 14.6K | Apache-2.0 | 16 hours ago |
| drumih/turbo-fieldfare | ★ 6.9K | Apache-2.0 | 4 days ago |
An edge-native MoE serving engine that runs large open-weight models like DeepSeek-V4 and Qwen3.6 on consumer GPUs by splitting work between GPU and CPU memory.
Swift and Metal runtime that runs Gemma 4 26B-A4B on Apple Silicon Macs with about 2 GB of RAM by streaming experts from SSD.