imp MCP Server
kekzl/imp
[](https://agentrank-ai.com/tool/kekzl--imp/?utm_source=badge&utm_medium=readme&utm_campaign=agentrank_badge) LLM server for one RTX 5090. Runs Qwen3.8-Flash-Next (512-expert MoE) on a single 32 GB card at 75-80 tok/s. Decodes 17-191 % faster than llama.cpp on the same models. Built for agents: tool calls, long context, many streams at once.
claude mcp add agentrank -- npx -y agentrank-mcp-server Overview
kekzl/imp is a Cuda agent tool licensed under MIT. LLM server for one RTX 5090. Runs Qwen3.8-Flash-Next (512-expert MoE) on a single 32 GB card at 75-80 tok/s. Decodes 17-191 % faster than llama.cpp on the same models. Built for agents: tool calls, long context, many streams at once.
Ranked #694 out of 24845 indexed tools.
In the top 3% of all indexed tools.
Actively maintained with commits in the last week.
Ecosystem
Score Breakdown
42 stars → early stage
Last commit today → actively maintained
309/309 issues closed → responsive maintainer
3 contributors → small team
No dependents → no downstream usage
Weights: Freshness 20% · Issue Health 20% · Dependents 22% · Stars 10% · Contributors 8% · How we score →
Matched Queries
Get the weekly AgentRank digest
Top movers, new tools, ecosystem insights — straight to your inbox.