MiniCPM5 - Just How Good Can a 1B Model Be?

In this video, I look at MiniCPM5 from OpenBMB. This 1B model is certainly punching above its weight for agentic use cases and is a good fit for many on device projects. Checkout the Evomap Grant Program: https://evomap.ai/api-grant?invite=CUUQ4NFR For More info about Evomap: https://evomap.ai/?utm_source=sam&utm_medium=youtube&utm_id=01&utm_content=integration GH: https://github.com/openbmb/minicpm HF Collection: https://huggingface.co/collections/openbmb/minicpm5 Twitter: https://x.com/Sam_Witteveen 🕵️ Interested in building LLM Agents? Fill out the form below Building LLM Agents Form: https://drp.li/dIMes 👨‍💻Github: https://github.com/samwit/llm-tutorials ⏱️Time Stamps: 00:00 Intro 02:53 MiniCPM5-1B 04:00 Artificial Analysis Benchmark 04:25 Model Information 05:02 MiniCPM5-1B HF 05:20 Datasets 07:15 Benchmark 10:20 MiniCPM Desk Pet Demo 11:36 Demo 17:30 Evals