Havuz

An agentic system that auto-optimizes LLM workloads on AMD GPUs.

A next-generation automatic algorithm design platform, making automated algorithm design more accessible and easier to use

基于麦当劳 MCP 的点单决策求解器：在预算、热量、钠约束下求解满足感最高的组合

Asks what to optimise and how to measure it, then runs experiments on its own branch.

optim-agent, Claude Code, Codex veya OpenCode kullanan coding agent’larla ölçülebilir sistem parametrelerini deneyerek optimize eden bir Python aracıdır.

Neden yükseliyor · 82 günde 939 yıldıza ulaşması erken dönemde görünürlük kazandığını gösteriyor; coding agent’ları hyperparameter optimization sürecine bağlama fikri de ilgi çekici. Ancak son 28 günde commit ve farklı yazar olmaması, güncel geliştirme temposuna dair temkin gerektiriyor.

Distributed 56M-parameter LLM inference across 3 ESP32-S3 boards via ESP-NOW , Split-PLE + KV cache, fully offline.

CTX - Context Runtime Engine for Coding Agents

Flash weight streaming for MLX: run massive models larger than your RAM on Apple Silicon.

Ultra-Fast Python Solutions: 110x Speed Optimization Guide 2026

Sub-millisecond query routing & cost-aware execution for GraphRAG architectures. Adaptive-AtomRAG is a highly optimized conditional routing

Optimize LLM agent harnesses through tunable Formulas to dynamically enhance the agent, and improve those Formulas with optimizers based on

TurboQuant MLX: Flagship high-performance model quantization for Apple Silicon. Squeeze maximum inference speed and memory efficiency out of

A dependency-free Python genetic-algorithm engine with pluggable fitness criteria — the reusable search core for an LLM/AI autoresearcher.

Hillclimbing on verifiable rewards. Use Claude Code and Codex to autonomously search for python programs that optimize a score you define. 🏔

Dynamic per-token early exit for LLM inference. Skip layers tokens don't need

Agent-powered iterative optimization: put an AI agent in a verified improvement loop.

Math Modeling AI Agent Flow | 数学建模全流程 AI 助手（国赛 CUMCM / 美赛 MCM-ICM）。四阶段流水线 Solver→Paper→Finalizer→Growth：解题认证、论文写作、提交终检、赛后复盘。防 AI 幻觉、防版本冲突，可安

Kiasu Miles — A Singapore credit card rewards optimization MCP. Real-time card matching, miles tracking, spend analysis.

Semantic cache strategy bench: replay recorded request traces through candidate caching policies and measure which one actually pays off.

A managed proxy gateway and compiler for AI agents. It takes noisy OpenAPI/MCP directories, compresses them in flat mode, saves tokens, and

Composable LLM decoding on the probability simplex: library and paper code.

MCP-compatible context and memory system: paged code navigation for AI assistants, plus a recorded, verified theory of what is known about t

Kata builds the best AI agent for a subnet through open competition — so anyone can mine that subnet with a proven, optimized agent.

CCO is an open-source system for managing, validating, and improving GPU.