Humanity's Last ExamVerified
38.5%
HLE (AA run)
Models / OpenAI
GPT-6 Luna
GPT-6 · Released 2026-09-22
OpenAI's 22 Sep 2026 Luna line for high-volume work. API list $0.10 / $0.50 per 1M tokens on short context, 1.05M context, 128K output, knowledge cutoff 18 May 2026. No AA or Vals ledger row on this page as of 23 Sep 2026.
Current OpenAI Luna line for high-volume extraction, routing, and everyday agent calls.
Composite57
Context1M
Input / 1M$0.10
Output / 1M$0.50
Knowledge cutoff2026-05-18
Output speed136 t/s
Statusflagship
Family profile
Best published score in each covered benchmark family.
59/100 avg
- Reasoning
- 39
- Coding
- 55
- Long context
- 83
3 tested benchmark families
Editor's note
Recorded from the OpenAI API model page and pricing page on 23 Sep 2026, plus the 22 Sep developer announcement. The announcement also opens Luna to Free and Go users in the desktop app. No independent AA or Vals row yet. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where GPT-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 | #11/ 184 | 83 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #25/ 97 | 55 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #33/ 195 | Artificial Analysis | 2026-10-12 | |
| Humanity's Last Exam | Reasoning | #50/ 186 | 39 | 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-6 Luna shows up, it's highlighted.
ReasoningGPT-6 Luna · 39
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingGPT-6 Luna · 55
- 174
- 273
- 370
Long contextGPT-6 Luna · 83
- 189
- 286
- 385
Newest receipts
4 recentLast 120 days
Benchmark rows added to the public ledger for GPT-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 | 136 t/s | Artificial Analysis | 4 |
| AA-LCR | Long context | 83.33% | Artificial Analysis | 18 |
| Humanity's Last Exam | Reasoning | 38.5% | Artificial Analysis | 18 |
| SciCode | Coding | 54.6% | Artificial Analysis | 18 |
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-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-6 Luna. Scores link to the original source; gaps mean no public row exists yet.
4 of 13 catalog benchmarks have a sourced row for GPT-6 Luna.
Reasoning
Coding
SciCodeVerified
54.6%
SciCode (AA run)
Long context
AA-LCRVerified
83.33%
AA-LCR (AA run)
Performance
AA output speedVerified
136 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- EU data residency now includes Flex and Batch.
- Public on the OpenAI API, Codex, and ChatGPT Work.
Sources
Where the numbers come from
- OpenAI API model page· checked 2026-09-29
- OpenAI API pricing· checked 2026-09-29
- OpenAI developer announcement· checked 2026-09-23
- Vercel AI Gateway changelog· 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-6 Luna?
GPT-6 Luna is a OpenAI model: OpenAI's 22 Sep 2026 Luna line for high-volume work. API list $0.10 / $0.50 per 1M tokens on short context, 1.05M context, 128K output, knowledge cutoff 18 May 2026. No AA or Vals ledger row on this page as of 23 Sep 2026.
What is GPT-6 Luna for?
Current OpenAI Luna line for high-volume extraction, routing, and everyday agent calls.
What does GPT-6 Luna cost?
GPT-6 Luna lists $0.10 input and $0.50 output per 1M tokens. Pricing last checked 2026-09-29.
How large is the GPT-6 Luna context window?
GPT-6 Luna publishes a 1M token context window.
What is the VerdictPal status of GPT-6 Luna?
GPT-6 Luna carries Flagship 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.