OpenAI Releases GPT-5.6 (Sol, Terra, Luna): A Three-Tier Model Family With Programmatic Tool Calling in the Responses API
OpenAI simply moved the GPT-5.6 family to common availability, following a restricted preview. The launch ships three fashions moderately than one. Sol is the flagship, Terra is the balanced on a regular basis tier, and Luna is the most cost-efficient.
To sum up
- GPT-5.6 ships three tiers — Sol, Terra, Luna — priced from $1/$6 to $5/$30 per 1M tokens.
- Sol leads the Artificial Analysis Coding Agent Index at 80, 2.8 factors above Claude Fable 5.
- Programmatic Tool Calling runs model-written JavaScript in an remoted V8 runtime with no community entry.
extremelyruns 4 brokers in parallel, lifting Terminal-Bench 2.1 from 88.8% to 91.9%.- SWE-Bench Pro stays a niche: Sol’s 64.6% trails Claude Mythos 5’s 80.3% by roughly 15 factors.
What is GPT-5.6?
Three fashions, one era, priced per 1M tokens. Sol is $5 enter and $30 output. Terra is $2.50 and $15. Luna is $1 and $6.
Availability differs by floor:
- Chat: Plus, Pro, Business, and Enterprise customers entry Sol at medium and better effort. Pro and Enterprise may also choose GPT-5.6 Sol Pro.
- ChatGPT Work and Codex: Free and Go customers entry Terra. Paid customers select amongst all three and set effort per mannequin.
maxis accessible to all customers with GPT-5.6 entry and is toggled in settings. - API: All three tiers can be found. Programmatic Tool Calling and a multi-agent beta each dwell in the Responses API.
Prompt caching additionally modified. GPT-5.6 helps specific cache breakpoints and a 30-minute minimal cache life. Cache writes are billed at 1.25x the mannequin’s uncached enter fee. Cache reads proceed to obtain the 90% cached-input low cost.
Performance
Furthermore, Agents’ Last Exam evaluates long-running skilled workflows throughout 55 fields. OpenAI studies a brand new excessive of 53.6 for Sol. It describes this as eclipsing Claude Fable 5 (adaptive reasoning) by 13.1 factors.
OpenAI’s personal eval desk lists Sol at 52.7% and Fable 5 at 40.5%. The 13.1-point hole matches 53.6 minus 40.5, so the Fable 5 baseline is constant throughout each. Only Sol’s determine differs. OpenAI doesn’t label which reasoning configuration produced 53.6.
On the Artificial Analysis Coding Agent Index v1.1, Sol at max reasoning scores 80. That is 2.8 factors above Fable 5. OpenAI studies it does so utilizing lower than half the output tokens and fewer than half the time.
Sol units new state-of-the-art outcomes on Terminal-Bench 2.1 and DeepSWE. It reaches 92.2% on BrowseComp and 62.6% on OSWorld 2.0. On OSWorld it surpasses Claude Opus 4.8 whereas utilizing 85% fewer output tokens.
| Eval | GPT-5.6 Sol | GPT-5.6 Terra | GPT-5.6 Luna | GPT-5.5 | Claude Fable 5 | Claude Opus 4.8 | Gemini 3.1 Pro Preview |
|---|---|---|---|---|---|---|---|
| AA Coding Agent Index v1.1 | 80 | 77.4 | 74.6 | 76.4 | 77.2 | 72.5 | 42.7 |
| AA Intelligence Index v4.1 | 58.9 | 55 | 51.2 | 54.8 | 59.9 | 55.7 | 46.5 |
| Terminal-Bench 2.1 | 88.8% | 87.4% | 84.7% | 85.6% | 83.1% | 78.9% | 70.7% |
| DeepSWE v1.1 | 72.7% | 69.6% | 67.2% | 67% | 69.7% | 59% | 11.8% |
| SWE-Bench Pro | 64.6% | 63.4% | 62.7% | 59.4% | 80% | 69.2% | 54.2% |
| Agents’ Last Exam | 52.7% | 50.4% | 50.3% | 46.9% | 40.5% | 45.2% | 32.1% |
| GDPval-AA v2 (Elo) | 1,747.8 | 1,593 | 1,591.8 | 1,493.7 | 1,759.6 | 1,600.1 | 962.3 |
| BrowseComp | 90.4% | 87.5% | 83.3% | 84.4% | — | 84.3% | 85.9% |
| OSWorld 2.0 | 62.6% | 50.2% | 45.6% | 47.5% | — | 54.8% | — |
| Toolathlon | 58% | 53.1% | 53.4% | 55.6% | 61.7% | 59.9% | 48.8% |
Where GPT-5.6 Does Not Lead
However, 4 gaps are price naming:
- SWE-Bench Pro: Sol scores 64.6%. Claude Mythos 5 scores 80.3% and Fable 5 scores 80%. That is a roughly 15-point deficit on a broadly watched coding eval.
- Broad intelligence and data work: Fable 5 leads the Artificial Analysis Intelligence Index v4.1, 59.9 to 58.9. Fable 5 additionally leads GDPval-AA v2 by about 12 Elo. On Well beingBench Professional, Fable 5 scores 60.9% towards Sol’s 60.5%.
- Tool use: On Toolathlon, Sol scores 58%. Fable 5 reaches 61.7% and Opus 4.8 reaches 59.9%. Luna additionally edges out Terra right here, inverting the tier order.
- Long context: Luna drops to 41.3% on OpenAI MRCR v2 8-needle, at each 256K–512K and 512K–1M. Sol scores 73.8% at 512K–1M, barely under GPT-5.5’s 74%.
Interactive Explainer
GPT‑5.5
Anthropic
Built by MARKTECHPOST
