Fit any LLM under a memory budget — an optimizer/planner that picks per-layer quantization, KV-cache precision, context length and GPU/CPU/disk offload to run large language models locally, then emits ready-to-run llama.cpp / HuggingFace / Ollama commands.
machine-learning ai quantization kv-cache model-optimization llm llama-cpp local-llm llm-inference gguf llm-optimizer ai-optimizer
-
Updated
May 30, 2026 - Python