gpustack-modelsync
Sync GPUStack model folders across cluster nodes via Syncthing — pick which models live on which nodes in a small matrix UI, with automatic register/deregister and stuck-folder recovery.
- 0
- Python
- MIT
- 2 wk ago
Sync GPUStack model folders across cluster nodes via Syncthing — pick which models live on which nodes in a small matrix UI, with automatic register/deregister and stuck-folder recovery.
Benchmark OpenAI-compatible LLM servers: find throughput peaks and latency SLO capacity, with terminal charts.
Manage heterogeneous accelerators, vGPU slicing, shared KV cache, model delivery, and inference workloads.