GPQA DiamondVerified
91.1%
GPQA Diamond (AA run)
Models / OpenAI
GPT-5.6 Luna
GPT-5.6 · Released 2026-07-09
OpenAI's generally available GPT-5.6 Luna line for high-volume work. API list for short context, checked 23 Sep 2026, is $0.20 / $1.20 per 1M tokens. GPT-6 Luna is the current Luna tier.
GPT-5.6 tier for high-volume sub-agent, extraction, routing, and everyday coding workloads.
Composite59
ContextUnlisted
Input / 1M$0.20
Output / 1M$1.2
Knowledge cutoffunpublished
Output speed120 t/s
Statussolid
Family profile
Best published score in each covered benchmark family.
78/100 avg
- Reasoning
- 91
- Coding
- 54
- Agentic
- 81
- Long context
- 84
4 tested benchmark families
Editor's note
Luna moved from limited preview to general availability on July 9, 2026. Independent benchmark rows now cover the family, while OpenAI still has not published the context limit.Benchmark placements
Per-benchmark positions
Where GPT-5.6 Luna places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| AA-LCR | Long context | #8/ 184 | 84 | Artificial Analysis | 2026-09-28 |
| Terminal-Bench-Science | Agentic | #9/ 9 | 3 | Terminal-Bench-Science 0.1 announcement | 2026-08-27 |
| Terminal-Bench | Agentic | #22/ 167 | 81 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #31/ 97 | 54 | Artificial Analysis | 2026-09-28 |
| GPQA Diamond | Reasoning | #35/ 184 | 91 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #45/ 195 | Artificial Analysis | 2026-10-12 | |
| Humanity's Last Exam | Reasoning | #46/ 186 | 40 | Artificial Analysis | 2026-09-28 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where GPT-5.6 Luna shows up, it's highlighted.
ReasoningGPT-5.6 Luna · 91
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingGPT-5.6 Luna · 54
- 174
- 273
- 370
AgenticGPT-5.6 Luna · 81
- 191
- 289
- 389
Long contextGPT-5.6 Luna · 84
- 189
- 286
- 385
Newest receipts
7 recentLast 120 days
Benchmark rows added to the public ledger for GPT-5.6 Luna in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 120 t/s | Artificial Analysis | 4 |
| AA-LCR | Long context | 83.67% | Artificial Analysis | 18 |
| GPQA Diamond | Reasoning | 91.1% | Artificial Analysis | 18 |
| Humanity's Last Exam | Reasoning | 39.5% | Artificial Analysis | 18 |
| SciCode | Coding | 53.6% | Artificial Analysis | 18 |
| Terminal-Bench | Agentic | 80.9% | Artificial Analysis | 18 |
| Terminal-Bench-Science | Agentic | 3.3% | Terminal-Bench-Science 0.1 announcement | 50 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where GPT-5.6 Luna sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for GPT-5.6 Luna. Scores link to the original source; gaps mean no public row exists yet.
7 of 13 catalog benchmarks have a sourced row for GPT-5.6 Luna.
Reasoning
Humanity's Last ExamVerified
39.5%
HLE (AA run)
Coding
SciCodeVerified
53.6%
SciCode (AA run)
Agentic
Terminal-BenchVerified
80.9%
Terminal-Bench v2.1 (AA run)
Terminal-Bench-ScienceVerified
3.3%
TB-Science 0.1 resolution (Codex)
Long context
AA-LCRVerified
83.67%
AA-LCR (AA run)
Performance
AA output speedVerified
120 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- API list re-checked. Standard for short context is $0.20 input and $1.20 output per 1M tokens.
- Recorded general availability across ChatGPT, Codex, and the API and refreshed benchmark evidence.
- Updated public pricing: Luna is listed at $1 input / $6 output per 1M tokens.
- Previewed as the fastest and most cost-efficient GPT-5.6 tier.
Sources
Where the numbers come from
- OpenAI GPT-5.6 launch· checked 2026-07-19
- OpenAI GPT-5.6 preview availability and pricing· checked 2026-07-19
- OpenAI GPT-5.6 preview system card· checked 2026-07-19
- OpenAI API pricing· checked 2026-09-23
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is GPT-5.6 Luna?
GPT-5.6 Luna is a OpenAI model: OpenAI's generally available GPT-5.6 Luna line for high-volume work. API list for short context, checked 23 Sep 2026, is $0.20 / $1.20 per 1M tokens. GPT-6 Luna is the current Luna tier.
What is GPT-5.6 Luna for?
GPT-5.6 tier for high-volume sub-agent, extraction, routing, and everyday coding workloads.
What does GPT-5.6 Luna cost?
GPT-5.6 Luna lists $0.20 input and $1.2 output per 1M tokens. Pricing last checked 2026-09-23.
How large is the GPT-5.6 Luna context window?
GPT-5.6 Luna does not publish a context window on the sources checked for this page.
What is the VerdictPal status of GPT-5.6 Luna?
GPT-5.6 Luna carries Solid status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.