Search 5 packages
| Packages1–5 of 5 | Version | Risk | Vulns | Description | Downloads | When |
|---|
| ironbark cargo | | medium 0.1.0 did not build | – | Run Maple (ternary MoE) on Apple Silicon with experts streamed from storage, sized to the device's memory | 10 | 4h ago |
| ironbark-fetch cargo | | – 0.1.0 did not build | – | Build a local experts.bin from a remote copy: on-demand records, hash checks, background fill | 12 | 4h ago |
| ironbark-experts cargo | | – 0.1.0 did not build | – | Event-driven MoE expert store: loader pool, per-layer cache over experts.bin | 18 | 4h ago |
| ramvamp-core cargo | 0.1.1 | none | none | Runtime for streaming fine-grained MoE experts from NVMe: packed format, io_uring reads, LFU expert cache, CPU kernels, KV cache, generation loop | 79 | 6d ago |
| rusting_brain cargo | 3.0.0 | none | none | A Rust deep-learning library for training and running transformer language models, dense and mixture-of-experts, on CPU or CUDA. | 122 | 11d ago |
