From CLIs to LLMs: 5 cool applications to build with GraalVM
summary
Shows five concrete apps built with GraalVM native image: a 40 MB Spring Shell CLI that starts instantly; a bundled full‑stack travel app calling OpenAI’s API; a single‑file Java LLM inference engine running Llama 3.2 1B locally and interactively; benchmarks report JVM ≈1.4k RPS, native image baseline ≈1.1k RPS, optimized native ≈1.5k RPS; covers PGO and ML‑PGO, vector API and FFM for fast inference, polyglot Python integration, and attack‑surface reduction by stripping unused code.
AI-written summary · about
chapters (8)
- 0:00Introduction and overview
- 1:49What is GraalVM Native Image
- 6:42CLI application with Spring Shell
- 9:51Full-stack travel app demo
- 14:04DIY LLM inference engine
- 19:28Benchmarking performance
- 35:50Polyglot Java and Python app
- 44:12GraalVM 25 preview and resources
related talks
47mFrom CLIs to LLMs: 5 Cool Applications to Build with GraalVM
35mTop 10 Things to Do with GraalVM
53mBootiful GraalVM
47mBring the action: using GraalVM in production
46mBring the Action: Using GraalVM in Production
45mGoing AOT: Everything you need to know about GraalVM for Java applications
50mGoing AOT: Everything you need to know about GraalVM for Java
49mGraalVM: Run Programs Faster Anywhere