GPT-5.6 Is Here
OpenAI publicly released GPT-5.6 on July 9, 2026 (after a phased, government-reviewed rollout that began June 26). It's not one model but a family of three tiers under a new naming scheme — and the coding numbers are the headline. Here's what's confirmed and how to start using it.
The New Naming: Sol, Terra, Luna
With GPT-5.6, the number is the generation and the name is the capability tier:
- Sol — the flagship/workhorse; OpenAI's strongest and most efficient model yet.
- Terra — the balanced middle tier; roughly GPT-5.5-level performance at about half the cost.
- Luna — the budget tier; OpenAI says it outperforms Claude Opus 4.8.
Pricing (per 1M tokens)
| Tier | Input | Output | Positioning |
|---|---|---|---|
| Sol | $5.00 | $30.00 | Flagship, SOTA coding |
| Terra | $2.50 | $15.00 | ~GPT-5.5 perf, 2x cheaper |
| Luna | $1.00 | $6.00 | Budget, beats Opus 4.8 |
GPT-5.6 also improves prompt caching: explicit cache breakpoints and a 30-minute minimum cache life, with cache reads keeping the ~90% discount (cache writes billed at 1.25x uncached input). Structure prompts to exploit it — see prompt caching guide.
What's Actually New
- Coding SOTA. OpenAI reports Sol scores 80 on the Artificial Analysis Coding Agent Index — ahead of Anthropic's Fable 5 — while using less than half the output tokens and ~1/3 less cost. Terra lands just above Fable 5; Luna above Opus 4.8.
- Cybersecurity. Billed as OpenAI's "strongest cybersecurity model yet" (the reason for the phased, government-reviewed launch).
- Design judgment & computer use. Stronger UI/design output and computer-use that inspects and refines rendered results, not just code.
- Efficiency. Altman claims Sol is ~54% more token-efficient on coding tasks — meaningful for agent bills.
How to Call GPT-5.6
GPT-5.6 is available via the OpenAI API and Codex. If you're routing across providers, you can call the GPT family through one key on LinkModel alongside Claude, Gemini and DeepSeek — see the model list:
curl -X POST https://api.linkmodel.ai/v1/chat/completions \
-H "Authorization: Bearer $LINKMODEL_API_KEY" -H "Content-Type: application/json" \
-d '{ "model": "gpt-5.6-terra", "messages": [{"role":"user","content":"Refactor this module and add tests: ..."}] }'Confirm the exact model strings in the docs. New to keys? See how to get an OpenAI API key.
Which Tier Should You Use?
- Hardest coding / agents → Sol (but $30 output — reserve for the hard 5%).
- Everyday production → Terra (GPT-5.5-level at half the price).
- High-volume / cheap → Luna ($1/$6, strong for its tier).
Route the bulk to Luna/Terra and escalate to Sol only when needed — the tiered pattern in how to reduce AI API costs. Compare against rivals in best coding LLM API and Claude vs GPT; full rates in GPT API pricing.
Bottom Line
GPT-5.6 splits into Sol (frontier coding), Terra (value default) and Luna (cheap, punchy). If you build coding agents, Sol's efficiency is the story; for everyday work, Terra is the sweet spot. Test it against Claude and Gemini on your own prompts before committing.
Start free with a $1 credit and benchmark GPT-5.6.