Orchestrator + Subagents in opencode with Open Models
Melvin Vivas · X post · 2026-09-11 · Open on X
Topics: AI Agents, Tool Use & MCP, AI Dev Tools & Productivity · Level: intermediate
Summary
Melvin Vivas is testing opencode to recreate the orchestrator-and-subagents pattern with open models such as DeepSeek, Qwen, GLM and Kimi. In his hybrid setup, cloud models do the orchestration and smaller local models run as subagents.
Key points
- The orchestrator-plus-subagents pattern can be built in opencode, not only in proprietary coding agents.
- Open models to try: DeepSeek, Qwen, GLM and Kimi.
- Hybrid design: a stronger cloud model orchestrates while locally running models act as subagents.
- Running subagents locally can cut API cost and keep some work on your own machine.
Resources mentioned
- OpenCode · tool · x.com · free
Open-source terminal coding agent that can be configured with different model providers, including open models.
Also in: OpenCode Console Security Breach: Check Your Provider Keys (Melvin Vivas on X · notes), DeepSeek V4.1 Flash Now Free in OpenCode (Melvin Vivas on X · notes), OpenAI Makes GPT-6 Astra and GPT-6.1 Sol About 50% Faster (Melvin Vivas on X · notes), T3Code: a better interface for terminal coding agents (Melvin Vivas on X · notes) and 13 more - DeepSeek · tool · deepseek.com · free
DeepSeek's updated Flash model, said to be faster with shorter outputs, available through DeepSeek's web app and API.
Also in: DeepSeek-V4.1-Flash-1017 release: better benchmarks, shorter outputs (Melvin Vivas on X · notes), Open models now dominate token volume on Vercel AI Gateway (Melvin Vivas on X · notes), jina-ocr-v1: Turning PDFs, Scans and Tables into Markdown (Melvin Vivas on X · notes), Bolt.new Adds Open Models (GLM, DeepSeek, Kimi) via Bolt Forge (Melvin Vivas on X · notes) and 5 more - Qwen · tool · github.com · free
Alibaba's family of open-weight language models, including small models suited to fine-tuning and running locally.
Also in: Meetily: Open-Source Local AI Meeting Notes App (Melvin Vivas on X · notes), Local AI vs Cloud AI: When Running Models Locally Makes Sense (Melvin Vivas on X · notes), Local AI vs Cloud AI: Using the Right Model for the Right Job (Melvin Vivas on X · notes), OpenCode Go: Open Coding Models (DeepSeek, Qwen, Kimi, GLM) for $10/mo (Melvin Vivas on X · notes) and 2 more - GLM · tool · github.com · free
Chinese open-weight LLM family from Zhipu AI.
Also in: OpenCode Go: Open Coding Models (DeepSeek, Qwen, Kimi, GLM) for $10/mo (Melvin Vivas on X · notes), Why Chinese AI Models Are Mostly Open Source (Melvin Vivas on X · notes) - Kimi · tool · github.com · free
Open-weight large language models from Moonshot AI, aimed at agent and coding tasks.
Also in: Bolt.new Adds Open Models (GLM, DeepSeek, Kimi) via Bolt Forge (Melvin Vivas on X · notes), OpenCode Go: Open Coding Models (DeepSeek, Qwen, Kimi, GLM) for $10/mo (Melvin Vivas on X · notes), Why Chinese AI Models Are Mostly Open Source (Melvin Vivas on X · notes)
Try this
- Set up opencode with a cloud orchestrator model and local open models (e.g. Qwen) as subagents.
More in AI Agents, Tool Use & MCP
- Claude Code mockup: a UI for managing agents in sandboxes
- Bionic by LM Studio: An Agent App for Local Models
- Updating Skills and AGENTS.md for Astra with OpenAI's Guidance
- Building Software Factories on the OpenAI Agents API (Codex-Powered)
- Codex Astra/Luna Orchestrator Repo: Orchestrator + Subagents
- GPT-Live-1 Voice Agents in the API, and Charlie Guo