diff --git a/llm-tools.md b/llm-tools.md
index d7d8583..4223012 100644
--- a/llm-tools.md
+++ b/llm-tools.md
@@ -973,6 +973,8 @@
 - [aisuite](https://github.com/andrewyng/aisuite) thin wrapper around python client libraries to seamlessly swap out different LLM providers without changing their code
 - [SCUDA](https://github.KeVmo314/Scuda) enables GPU-over-IP connectivity, allowing remote GPUs to be utilized by CPU-only machines for accelerated computing tasks such as model training and inference.  
 - [BitNet](https://github.com/microsoft/BitNet) bitnet.cpp is the official inference framework based on llama.cpp to run 1-Bit LLMs for fast and lossless inference of 1.58-bit models on CPU, NPU and GPU
+- [Paralleliq](https://paralleliq.ai) model-aware GPU fleet optimization layer for AI inference clusters. Detects tier misplacement, idle capacity, KV cache pressure, and throughput suppression with human-approval workflows and audit trails.
+- [piqc](https://github.com/paralleliq/piqc) open-source, read-only GPU waste scanner for Kubernetes inference clusters. Deploys in minutes, no write permissions required.
 
 ### Optimization