Testing Model Combinations for Codex Orchestration

Melvin Vivas · X post · 2026-09-06 · Open on X

Topics: AI Agents, Tool Use & MCP, AI Dev Tools & Productivity · Level: intermediate

Summary

The creator plans to test his Astra orchestrator + Luna subagents setup more thoroughly. He also wants to try other model mixes, such as adding Terra, to make Codex more efficient. The Plus plan's 5-hour usage limits are slowing his testing.

Key points

• Goal: make coding with Codex more efficient by mixing models across orchestrator and subagent roles.
• Next step is trying combinations that add the Terra model.
• The Plus plan's 5-hour usage limits make heavy multi-agent testing hard.

Resources mentioned

• donvito/codex-astra-luna-orchestrator · repo · github.com · free
  The creator's Codex skill for an orchestrator-plus-subagents setup (Astra orchestrator, Luna subagents), configured through config.toml.
  Also in: Codex Orchestrator Pattern: Astra/Sol Orchestrator with Luna Subagents (Melvin Vivas on X · notes), Open-source Codex skill: Astra/Sol orchestrator with Luna subagents (Melvin Vivas on X · notes), Cost-Efficient Codex Subagents: Astra/Sol Orchestrator + Luna Workers (Melvin Vivas on X · notes), Codex Orchestrator v0.2.1: GPT-6 Sol orchestrator with Luna subagents (Melvin Vivas on X · notes) and 27 more

Try this

• Compare cost and quality of different orchestrator and subagent model combinations in Codex.