Measure Your Own Codex Token Burn: Orchestrator+Subagents vs Single Agent
Melvin Vivas · X post · 2026-09-10 · Open on X
Topics: AI Dev Tools & Productivity, AI Agents, Tool Use & MCP, LLMOps, Deployment & Monitoring · Level: intermediate
Summary
Melvin Vivas logged his Codex usage and asked ChatGPT to analyze it. For him, an Astra orchestrator with Luna subagents was cheaper than a single Astra agent at high/xhigh. His main lesson is to take your own measurements, because the two setups were used on different projects.
Key points
- Log your own token usage and burn rate, then have an LLM such as ChatGPT analyze it.
- For him, Astra (orchestrator) + Luna (subagents) was cheaper.
- The comparison isn't apples to apples: the orchestrator setup was a greenfield build (local-evals), while Astra high/xhigh was used to add features to an existing app (Coworker).
- Pick the workflow that your own measurements support; it's your tokens.
Resources mentioned
- donvito/local-evals · repo · github.com · free
Eval app that runs locally and tests local models or any OpenAI-compatible API on JSON extraction and tool calling.
Also in: Local Evals v0.1.0: Run LLM Evals Locally (Melvin Vivas on X · notes), local-evals: A Local Eval App with a Setup Wizard (Melvin Vivas on X · notes), Melvin Vivas's open-source AI tools, built with Codex (Melvin Vivas on X · notes), local-evals: A Local LLM Eval App Built by Codex Subagents (Melvin Vivas on X · notes) and 1 more - Coworker (donvito/coworker) · repo · github.com · free
The creator's agent/bot tool. Version 0.7.0 adds a Primary Bot, free web search, an activity dashboard and support for open models through OpenRouter.
Also in: Coworker: a local-first desktop AI coworker that works with any model (Melvin Vivas on X · notes), Liquid AI's LFM2.5-24B-A2B: a local MoE model for agent tasks (Melvin Vivas on X · notes), Coworker: A Desktop AI Assistant App, Not a Coding Agent (Melvin Vivas on X · notes), Coworker v0.7.2: Open-Source Desktop AI Assistant (Melvin Vivas on X · notes) and 59 more - ChatGPT · tool · chatgpt.com · free · recommended by both Bashiri Smith & Melvin Vivas
OpenAI's AI assistant (web, desktop and mobile apps), used for chat, coding, writing and image generation.
Also in: Keep a Change Log When Multiple AI Agents Edit One Notion Doc (Melvin Vivas on X · notes), Multi-Agent Planning in Notion: Grok Bot, ChatGPT and Claude (Melvin Vivas on X · notes), Hark Pro Launch: A Proactive Personal AI Agent with Cloud Computer Use (Melvin Vivas on X · notes), Why Grok Bot Could Be the Next "ChatGPT Moment" for Personal AI Assistants (Melvin Vivas on X · notes) and 35 more
Try this
- Log your Codex usage (burn rate) for each setup and compare them yourself before picking a workflow.
More in AI Dev Tools & Productivity
- Use Codex GPT-5.3 Spark for Testing and Small Changes
- Making Product Demo Videos with Codex: The Coworker Desktop App Demo
- Codex Astra Making a Demo Video with HyperFrames
- Single-Agent Codex Astra xhigh Usage for a Feature Build
- Measuring Codex Usage Burn Rate: Astra (high) Without Subagents
- Editing a GitHub Repo From ChatGPT Work on iOS