Urgent.News

What's breaking now, across thousands of outlets.

AI

รัน local LLM ในแอป ChatGPT Desktop ด้วย opencodex ตัวเดียว

รัน local LLM ในแอป ChatGPT Desktop ด้วย opencodex ตัวเดียว โดย Nokka (นก-กา) | 19 กันยายน 2026 บทความนี้เขียนโดย AI (โมเดล deepseek-v4.1-flash ของผู้ให้บริการ ollama-cloud) ผ่าน Hermes Agent จาก Nous Research ตรวจสอบและเรียบเรียงโดย Nokka ข้อความในเครื่องหมายคำพูดที่เป็นคำแปลเป็นคำแปลของผม ไม่ใช่สำเนาต้นฉบับ มีเครื่องมือตัวหนึ่งที่ทำให้แอป ChatGPT Desktop คุยกับโมเดลที่ไม่ใช่ของ OpenAI ได้…

A local Large Language Model (LLM) can now be run within the ChatGPT Desktop app using a tool called opencodex. Developed by Nokka, this open-source solution works as a proxy on your own computer, allowing you to interact with models like MiniMax M3, Qwen 3.8 27B, and GLM-5.3-Flash, rather than just OpenAI's models. The opencodex tool modifies the base URL in the app to point to a local proxy, enabling it to communicate with various models without requiring a fork or modification of the original app.

This proxy service is MIT licensed and can work alongside the original Codex Desktop app, allowing for flexibility in choosing which models to use. The tool is easy to install and use, and it supports multiple runtimes, including Ollama, vLLM, and LM Studio.

Brief written by urgent.news from Dev.to's own syndicated text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Build and Test an AI Agent Skill with SKILL.md and Python

AI agents are good at interpreting intent, but they are not a reliable place to hide every rule in a workflow. If an agent must count characters, parse a file, or refuse to overwrite an existing…

  • Create skill structure with SKILL.md and scripts directory
  • Implement deterministic script to validate Conventional Commit messages
  • Test script thoroughly covering various message scenarios

The More Powerful the AI, the More the Architecture Matters

The boundaries I designed, the gaps I haven't solved, and why the difference matters Part 14 findings of an experiment: building an LLM-powered support agent with deterministic boundaries.

  • AI architecture crucial for effectiveness, as shown in experiment
  • Deterministic boundaries, allowlist, and gate prevent unexpected dependencies
  • Gate ensures low-risk tasks handled by AI, high-risk tasks require human oversight

More from Saturday 19 September →