Tokenflood – simulate arbitrary loads on instruction-tuned LLMs
Hi everyone, I just released an open source load testing tool for LLMs: https://github.com/twerkmeister/tokenflood === What is it and what problems does it solve? === Tokenflood is a load testing tool for instruction-tuned LLMs hat can simulate arbitrary LLM loads in terms of prompt, prefix, and output lengths and requests per second. Instead of first collecting prompt data for different load types, you can configure the desired parameters for your load test and you are good to go. It also let's you assess the latency effects of potential prompt parameter changes before…
In plain words
Tokenflood is an open-source load testing tool designed for instruction-tuned large language models. It allows developers to simulate arbitrary loads by configuring prompt length, prefix length, output length, and requests per second without needing to collect real prompt data first. The tool helps teams building latency-sensitive LLM applications test self-hosted model setups and evaluate how changes to prompt parameters would affect performance before implementation.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi everyone, I just released an open source load testing tool for LLMs: https://github.com/twerkmeister/tokenflood === What is it and what problems does it solve? === Tokenflood is a load testing tool for instruction-tuned LLMs hat can simulate arbitrary LLM loads in terms of prompt, prefix, and output lengths and requests per second. Instead of first collecting prompt data for different load types, you can configure the desired parameters for your load test and you are good to go. It also let's you assess the latency effects of potential prompt parameter changes before spending the time and effort to implement them. I believe it's really useful for developing latency sensitive LLM applications and * load testing self-hosted LLM model setups * Assessing the latency benefit of changes to prompt parameters before implementing those changes * Assessing latency and intraday variation of latency on hosted LLM services before sending your traffic there === Why did I built it? === Over the course of the past year, part of my work has been helping my clients to meet their latency, throughput and cost targets for LLMs (PTUs, anyone? ). That process involved making numerous choices about cloud providers, hardware, inference software, models, configurations and prompt changes. During that time I found myself doing similar tests over and over with a collection of adhoc scripts. I finally had some time on my hands and wanted to properly put it together in one tool. === What am I looking for? === I am sharing this for three reasons: Hoping this can make other's work for latency-sensitive LLM applications simpler, learning and improving from feedback, and finding new projects to work on. So please check it out on github (https://github.com/twerkmeister/tokenflood), comment, and reach out at [email protected] or on linkedin(https://www.linkedin.com/in/twerkmeister/) for professional inquiries. === Pics === image of cli interface: https://github.com/twerkmeister/tokenflood/blob/main/images/... result image: https://github.com/twerkmeister/tokenflood/blob/main/images/...
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, November 2025
the whole month →
- IB
Life & fun · Nov 2025 · bitsnpieces.dev
- BBoing▲782
Life & fun · Nov 2025 · boing.greg.technology