Pinned Loading

1. llm-router llm-router Public

   Cache-aware router for OpenAI-compatible LLM servers, in Go. Per-worker radix trees route each request to the worker holding its KV prefix. Validated on 4x A100 + vLLM and Apple Silicon + llama.cpp.

   Go 4 1

2. distill-sql distill-sql Public

   On-device text-to-SQL distilled from GPT-4o-mini into Qwen2.5 (0.5B → 3B locally on M1 via mlx-lm LoRA, 7B+ on cloud A100). 847 MB at 62.5% on Spider dev; 3B variant hits 72.6%, 7B variant hits 75.0%.

   Python 1

3. folio-writer folio-writer Public

   Multi-agent article generator: Spring AI Alibaba StateGraph, token streaming over SSE, parallel image generation across six providers, Vue 3 frontend.

   Java 2

4. gpu-k8s-operator gpu-k8s-operator Public

   Kubernetes operator for rolling-window GPU-hour budgets. Stateless accounting (recomputes from API state every reconcile, 0.996 accuracy after operator kill). Enforcement via eviction, pause, or al…

   Go