OpenAI model roles laid out: Astra for hard workflows, Luna for repeatable tasks
gabrielchua · x · 2026-09-12
In his hackathon tips thread, gabrielchua laid out OpenAI's model lineup and when to use each, directly in Codex or your own apps and agents:
- ⭐️ Astra: your hardest end-to-end workflows
- ☀️ Sol: deeper analysis, judgment, and polish
- 🌏 Terra: everyday coding and tool use
- 🌕 Luna: clear, repeatable tasks
The breakdown aligns with Astra being positioned as the flagship, consistent with the 'GPT-6 Astra' fluid 3D graphics demo shown at the Astra Commons demo night.
More from coding & agent
- Models already understand harnesses: a minimal agent harness takes hours, but customization still needs you — omarsar0 · 2026-10-31
- Testing agents with explicit rejection records: a cheaper path to verified troubleshooting answers — krishnan · 2026-10-31
- krishnan proposes logging ruled-out troubleshooting hypotheses for future reopening — krishnan · 2026-10-31
- krishnan on autonomous lab agents: criticism changed the next search, not just verdicts — krishnan · 2026-10-31
- Multi-agent AI lab autonomously researches quantum spins in diamond, revises drifting experiments — krishnan · 2026-10-31
- Next 9 proposals explored 7 new directions: rejection became information — krishnan · 2026-10-31