Local AI coding: JetBrains Mellum 2.1, Junie on RTX Spark, OpenCode with Ollama
JetBrains has released Mellum 2.1 Thinking, an Apache-licensed coding model that can run locally, and its Junie tool can now run locally on NVIDIA RTX Spark, with RTX Spark laptops launching on 16 October. A guide shows how to set up OpenCode with Ollama locally.
Who is affected and how
- senior developer
- Try the local model on routine tasks and compare its quality with cloud-based tools.
- startup founder
- Running locally can cut API costs, but it comes with hardware costs.
Sources
- JetBrains releases Mellum 2.1 Thinking, a small Apache-licensed coding model built to run locally — Local AIlocal-ai.net · Oct 12, 2026
- Junie Local on NVIDIA RTX Spark: Free Local AI Codingblog.jetbrains.com · Oct 16, 2026
- JetBrains' local AI programming tool Junie Local has been adapted for NVIDIA RTX Spark at Day 0: the throughput is up to 55 word elements… — 哲见 · WISE AIspke.com · Oct 17, 2026
- RTX Spark: software updates for local AI - Techkalauztechkalauz.hu · Oct 15, 2026 · magyar
- NVIDIA RTX Spark: AI coding, testing and deployment on a single machine · BPDATA AI Newsnews.bpdata.com · Oct 16, 2026 · magyar
- OpenCode Ollama Setup: Run a Local AI Coding Agent - DEV Communitydev.to · Oct 12, 2026
- Pi Coding Agent: Setup, Local Models and Comparison (2026)vdf.ai · Oct 14, 2026
- What AI coding models really cost, 29 models benchmarked | Bitobito.ai · Oct 15, 2026
Machine-generated summary based on the titles and excerpts of the source articles. The facts come from the linked article.