Continuous-eval – Granular evaluation of GenAI pipelines
Hi HN - we are the creators of “continuous-eval”, an open-source tool to test and evaluate generative AI apps. "Continuous-eval" came from our efforts to measure, validate and improve the reliability of a finance AI copilot we were developing for banks. End-to-end evaluation was not enough for us. We wanted to have granular evaluations that help pinpoint the bottlenecks and identify what / how to improve. We’ve since developed more metrics and made the framework more flexible so it can evaluate components like agent tool use, code change, retrieval steps, etc. Let us know what you think…
In plain words
Continuous-eval is an open-source testing framework for evaluating generative AI applications at a granular level. It enables developers to assess individual components such as agent tool use, code changes, and retrieval steps rather than relying solely on end-to-end evaluation. The tool was created to help identify performance bottlenecks and guide improvements in AI systems, originally developed for finance applications. It is designed for teams building and refining generative AI products.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi HN - we are the creators of “continuous-eval”, an open-source tool to test and evaluate generative AI apps. "Continuous-eval" came from our efforts to measure, validate and improve the reliability of a finance AI copilot we were developing for banks. End-to-end evaluation was not enough for us. We wanted to have granular evaluations that help pinpoint the bottlenecks and identify what / how to improve. We’ve since developed more metrics and made the framework more flexible so it can evaluate components like agent tool use, code change, retrieval steps, etc. Let us know what you think of our approach to GenAI App evaluation.
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 · 26d ago · cactuscompute.com
Launched alongside, February 2024
the whole month →
MarketingIdeas.com▲1,230Never run out of marketing ideas ever again
Growth · 2024 · marketingideas.com
