Nexa MCP Server
nghiphaam/Nexa
Nexa: An advanced, lightweight language model featuring a built-in reasoning engine, agentic tool calling, and long-term vector memory. Optimized for high-efficiency inference.
claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
nghiphaam/Nexa is a Python agent tool licensed under Apache-2.0. Nexa: An advanced, lightweight language model featuring a built-in reasoning engine, agentic tool calling, and long-term vector memory. Optimized for high-efficiency inference.
Ranked #41 out of 109 indexed tools.
Actively maintained with commits in the last week.
Ecosystem
Score Breakdown
1 stars → early stage
Last commit today → actively maintained
No issues filed → no history to score
1 contributor → solo project
No dependents → no downstream usage
76 weekly installs → early adoption
Weights: Freshness 25% · Issue Health 25% · Dependents 25% · Stars 15% · Contributors 10% · How we score →
How to Improve
Badge all embed codes →
[](https://agentrank-ai.com/tool/nghiphaam--Nexa/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) <a href="https://agentrank-ai.com/tool/nghiphaam--Nexa/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge"><img src="https://agentrank-ai.com/api/badge/tool/nghiphaam--Nexa" alt="AgentRank"></a> Matched Queries
From the README
# Nexa 🌌 Nexa is an advanced, lightweight, open-source language model designed for high efficiency, flexible experimentation, and modern generative AI architectures. Grounded in a highly optimized codebase, Nexa brings modern architectural advancements alongside a full-stack reasoning engine. ## 🌟 Key Features ### 🏗 Architecture - **Modern Transformer Backbone:** Utilizes Grouped-Query Attention (GQA), SwiGLU Feed-Forward Networks (FFN), RMSNorm, and Rotary Position Embeddings (RoPE). - **Inference Optimizations:** Features an efficient KV Cache with precise temporal slot tracking, memory-state injections, and Speculative Decoding for accelerated local inference. ### 🧠 Reasoning & Agent Engine - **Internal Reasoning Engine:** An autonomous reasoning loop built directly into the chat runtime that orchestrates a **Planner**, a **Critic model**, and multi-path thought exploration to robustly verify and improve its own answers. - **Agent Tool Calling:** Embedded Python Sandbox mechRead full README on GitHub →
Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.