GPT-6 Luna
Made by OpenAI, released 22 Sep 2026. Epoch AI lists its access as “API access”. Appears solid at maths among the models listed here.
In short
What it does well
- Maths: Strong (15th of 43 listed models)
Where it falls short
- Coding: Behind the leaders (73rd of 77 listed models)
- Writing: Behind the leaders (73rd of 77 listed models)
- Knowledge: Middle (31st of 43 listed models)
What we don't know yet
- Fixing real software bugs: Not enough data
The four ratings
- Top tier
- Strong
- Middle
- Behind the leaders
- Not enough data
Ratings compare only the 77 models listed here, in quarters, best first. “Not enough data” means no result yet. It is never a low score.
- Writing
- Behind the leadersHow often people preferred its answers to writing prompts. 1,362 on Arena's scale · Source for GPT-6 Luna writing score (lmarena)
- Coding
- Behind the leadersHow often people preferred its answers to coding questions. 1,437 on Arena's scale · Source for GPT-6 Luna coding score (lmarena)
- Maths
- StrongHow many very hard maths problems it solved. 78.9% · Source for GPT-6 Luna maths score (epoch)
- Knowledge
- MiddleHow many short factual questions it answered correctly. 41.4% · Source for GPT-6 Luna knowledge score (epoch)
Evidence completeness: Broad evidence, 12 of 13 tests have a result.
Where it is relatively stronger or weaker
This compares each kind of request with the model's own overall position among the 77 listed models, using Arena's votes. It is about performance on different languages, kinds of request and topics, not about political or social bias.
Relatively stronger
No category stands out as clearly stronger than its overall position.
Relatively weaker
No category stands out as clearly weaker than its overall position.
Show every category we can see
| Category | Position | Arena score | Votes | Source |
|---|---|---|---|---|
| Expert-level prompts Kind of request | 65th of 77 | 1,441 95% range 1,434 to 1,448 | 8,948 | Source for Expert-level prompts |
| Hard prompts Kind of request | 75th of 77 | 1,408 95% range 1,403 to 1,413 | 53,299 | Source for Hard prompts |
| Long prompts Kind of request | 73rd of 77 | 1,407 95% range 1,401 to 1,412 | 36,874 | Source for Long prompts |
| Longer conversations Kind of request | 76th of 77 | 1,393 95% range 1,386 to 1,400 | 12,512 | Source for Longer conversations |
| Chinese Language | 74th of 76 | 1,430 95% range 1,421 to 1,438 | 7,417 | Source for Chinese |
| English prompts Language | 77th of 77 | 1,394 95% range 1,389 to 1,400 | 33,153 | Source for English prompts |
| French Language | 71st of 72 | 1,406 95% range 1,393 to 1,418 | 3,684 | Source for French |
| German Language | 55th of 55 | 1,378 95% range 1,360 to 1,397 | 1,283 | Source for German |
| Japanese Language | 51st of 54 | 1,368 95% range 1,350 to 1,386 | 1,743 | Source for Japanese |
| Korean Language | 52nd of 54 | 1,356 95% range 1,337 to 1,375 | 1,327 | Source for Korean |
| Non-English prompts Language | 77th of 77 | 1,380 95% range 1,375 to 1,385 | 53,254 | Source for Non-English prompts |
| Russian Language | 75th of 77 | 1,390 95% range 1,383 to 1,397 | 11,809 | Source for Russian |
| Spanish Language | 68th of 70 | 1,391 95% range 1,379 to 1,403 | 3,311 | Source for Spanish |
| Entertainment, sport and media Topic | 74th of 77 | 1,362 95% range 1,356 to 1,368 | 22,445 | Source for Entertainment, sport and media |
| Law and government Topic | 73rd of 76 | 1,396 95% range 1,387 to 1,404 | 6,888 | Source for Law and government |
| Medicine and health Topic | 75th of 75 | 1,355 95% range 1,346 to 1,364 | 6,079 | Source for Medicine and health |
| Science Topic | 77th of 77 | 1,387 95% range 1,381 to 1,394 | 14,249 | Source for Science |
| Software and IT Topic | 74th of 77 | 1,425 95% range 1,420 to 1,431 | 30,405 | Source for Software and IT |
Bias: what we can and can't see
Our sources do not measure political lean, stereotyping, flattery (agreeing with you too readily) or safety behaviour, so we cannot say how any model does on those. What we can show is above: where a model is relatively stronger or weaker across languages, kinds of request and topics. Also, people voting in Arena can favour answer style, such as length and formatting, and our numbers do not adjust for that. See what AI bias means and what these tests can tell you.
Best-supported use cases
This model is not among the options we recommend for any of the five tasks. That reflects the rules and the models listed, not a verdict on the model.
Where an alternative might be more suitable
- Writing and creating content: rated Behind the leaders on writing votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Coding and building things: rated Behind the leaders on coding votes. Compare it with Claude Opus 5.5, the best-supported choice for this task.
- Homework and maths: rated Strong on hard maths problems. Compare it with GPT-6.1 Sol, the best-supported choice for this task.
- Business emails and everyday work: rated Behind the leaders on business-task votes. Compare it with Gemini 4 Argon, the best-supported choice for this task.
- Research and checking facts: rated Middle on short factual questions. Compare it with Gemini 4 Argon, the best-supported choice for this task.
What it is
- Maker
- OpenAI
- Release date (Epoch AI)
- 22 Sep 2026
- Access, as listed by Epoch AI
- API access
- Licence, as listed by Arena
- Proprietary
- Maker's country (Epoch AI)
- United States of America
- Last observation in our data
- 28 Oct 2026
- Data last changed
- 29 Oct 2026, 14:36 UTC
We have not verified an official product page for this model, so we do not link to one.
How you can use it
Proprietary. The model's files are not published, so it is generally available only through its maker or partners the maker chooses.
How you can reach it (an app, a website, a developer API or a third-party tool) and what that costs are not covered by our benchmark data. Where we could verify prices from the maker's own pages, they are shown under Price. Epoch AI lists access as “API access”. See models, apps and harnesses and how AI pricing works.
Price
Verified consumer plans that include this model
- ChatGPT Free: GBP 0.00 per month. Maker's page, checked 29 Oct 2026. Comparison table cell for this model on this plan: "Yes". Limits apply.
- ChatGPT Go: GBP 7.00 per month. Maker's page, checked 29 Oct 2026. Comparison table cell for this model on this plan: "Yes". The page says this plan may include ads.
- ChatGPT Plus: GBP 20.00 per month. Maker's page, checked 29 Oct 2026. Comparison table cell for this model on this plan: "Yes". The page says limits apply.
- ChatGPT Pro (from): GBP 89.00 per month. Maker's page, checked 29 Oct 2026. Comparison table cell for this model on this plan: "Expanded". Listed as 'From £89 / month' with a choice of 3 usage tiers.
Prices change and can differ by country. Check the maker's page before you pay.
See ChatGPT plans and pricing (checked 29 Oct 2026). Page opened and read in the browser. Which plan includes which model comes from its comparison table.
We never infer subscription prices from API prices. New to “tokens” and “per million”? See how AI pricing works.
Good to remember
- Benchmarks are samples of tasks. None proves performance in every situation.
- Arena scores come from people choosing between answers. That measures preference, not correctness.
- Ratings are relative to the 77 models listed here. A different set would change them.
- Sources test models at different settings. Each result shows the setting used, and we do not adjust it.
The evidence, for the curious
Show all 13 tests with settings and sources
| Test | Result | Rating | Setting used | Source identifier | Observed | Links |
|---|---|---|---|---|---|---|
| Arena text: overall preference Arena 'text' leaderboard (Bradley-Terry rating, no style control) |
1,389 Arena score 95% range 1,385 to 1,393, 86,478 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: overall preference score (lmarena) |
| Arena text: creative writing Arena 'text' leaderboard, category creative_writing |
1,362 Arena score 95% range 1,356 to 1,368, 19,197 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: creative writing score (lmarena) |
| Arena text: coding Arena 'text' leaderboard, category coding |
1,437 Arena score 95% range 1,431 to 1,442, 19,790 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: coding score (lmarena) |
| Arena text: maths Arena 'text' leaderboard, category math |
1,410 Arena score 95% range 1,400 to 1,419, 4,768 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: maths score (lmarena) |
| Arena text: following instructions Arena 'text' leaderboard, category instruction_following |
1,406 Arena score 95% range 1,400 to 1,411, 29,447 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: following instructions score (lmarena) |
| Arena text: business and management Arena 'text' leaderboard, category industry_business_and_management_and_financial_operations |
1,384 Arena score 95% range 1,378 to 1,391, 14,761 votes |
Behind the leaders | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · text/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena text: business and management score (lmarena) |
| Arena WebDev: building web apps Arena 'webdev' leaderboard (Code Arena), category overall |
1,579 Arena score 95% range 1,574 to 1,585, 27,808 votes |
Strong | effort: Max, bradley-terry rating | GPT-6 Luna (Max) lmarena · webdev/latest |
28 Oct 2026 | Source for GPT-6 Luna Arena WebDev: building web apps score (lmarena) |
| FrontierMath Tiers 1-3 (v2) FrontierMath Tiers 1-3 v2 (private set), benchmark released 2026-06-12 |
78.9% share of problems solved standard error ±2.4 points |
Strong | effort: max, best score across scorers | gpt-6-luna_max epoch · frontiermath_tiers_1_3_v2.csv |
22 Sep 2026 | Source for GPT-6 Luna FrontierMath Tiers 1-3 (v2) score (epoch) No public run log in the data file. |
| OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025, benchmark released 2024-12-19 |
98.9% share of problems solved standard error ±0.7 points |
Strong | effort: max, best score across scorers | gpt-6-luna_max epoch · otis_mock_aime_2024_2025.csv |
22 Sep 2026 | Source for GPT-6 Luna OTIS Mock AIME 2024-2025 score (epoch) No public run log in the data file. |
| GPQA Diamond GPQA Diamond, benchmark released 2023-11-20 |
90.5% share of questions correct standard error ±1.9 points |
Middle | effort: max, best score across scorers | gpt-6-luna_max epoch · gpqa_diamond.csv |
22 Sep 2026 | Source for GPT-6 Luna GPQA Diamond score (epoch) No public run log in the data file. |
| SimpleQA Verified SimpleQA Verified, benchmark released 2025-09-09 |
41.4% share of questions correct standard error ±1.6 points |
Middle | effort: max, best score across scorers | gpt-6-luna_max epoch · simpleqa_verified.csv |
22 Sep 2026 | Source for GPT-6 Luna SimpleQA Verified score (epoch) No public run log in the data file. |
| SWE-bench Verified (Epoch runs) | Not enough data | |||||
| FrontierCode (reported by Cognition) FrontierCode, benchmark released 2026-06-08, aggregation Mean@5 |
42.4% mean score over 5 attempts no uncertainty figure published |
Not rated (results use different tools) | effort: max, tool: codex, mean@5 | gpt-6-luna_max epoch · frontiercode_external.csv |
not provided | Source for GPT-6 Luna FrontierCode (reported by Cognition) score (epoch) · Run details for FrontierCode (reported by Cognition) |
Missing evidence
Not enough data for these tests:
- SWE-bench Verified (Epoch runs)
Missing results are not low scores and do not count against the model.
Where a source gives no public run log, we cannot show the run behind the score. See data sources.
Not sure this is the right one?
Tell us what you need help with and we'll narrow the options down.