Models / GPT-6 Luna

OpenAI

GPT-6 Luna

GPT-6 Luna is best for marketing planning, and reasoning and knowledge; a strong pick at a modest price. The cheapest model of similar intelligence is MiMo-V2.6-Flash, at $0.18 per million tokens against $0.20.

Results 5 Nov 2026 · 121 results · 9 sources

Price per million tokens
$0.10 in · $0.50 out
OpenAI's list price, 5 Nov 2026 · compare prices$0.01 cached input · 50% off in batch
Speed
127 tokens a second; on its test prompt the answer starts after 116.0 s, thinking time included
Artificial Analysis, on its usual API, at max reasoning; 112–127 across 5 settings
Intelligence Index
38.1
Artificial Analysis, at max reasoning; 18.5–38.1 across 6 settings
Spring Prompt overall
21st of 31 score 39 of 100 · how it works
Developer
OpenAI, based in the US
Weights not published
Released
22 Sep 2026
1,050,000 tokens of context

Where it stands among the models you could choose

Available models grouped by price, intelligence, speed and developer location. Retired models are excluded.

AmongIntelligence IndexSpring Prompt overallPrice, blendedSpeed
Models available today 37th of 188Claude Opus 5.5 21st of 31Claude Opus 5.5 $0.20 blended, 18th percentilecheapest: Mistral Nemo 31st of 65Nemotron 3.5 Super VL
Models priced under $0.25 per million tokens 3rd of 42Claude Haiku 5.5 2nd of 3Claude Haiku 5.5 sets this group 9th of 19Nemotron 3.5 Super VL
Models with similar intelligence sets this group 3rd of 4Gemini 3.8 Flash $0.20 blended, 10th percentilecheapest: MiMo-V2.6-Flash 6th of 11DeepSeek-V4.1-Flash
Models writing 100–200 tokens a second 7th of 21Claude Sonnet 5.5 6th of 7Muse Wonder $0.20 blended, 10th percentilecheapest: Qwen3.5-9B sets this group
Models from developers based in the US 27th of 93Claude Opus 5.5 15th of 18Claude Opus 5.5 $0.20 blended, 19th percentilecheapest: Gemma 3 4B 20th of 33Nemotron 3.5 Super VL

The name under each rank is the leader of that group; green is the top quarter of the group and red the bottom quarter, by rank. Price is shown as a percentile: the share of the group that costs less, so lower is cheaper. "Sets this group" marks the measure the group is defined by. Spring Prompt overall is our score out of 100 across every source (how it works). Price is the developer's list price (or the typical OpenRouter host where we haven't read one) for three input tokens to one output; intelligence and speed are from Artificial Analysis (data sourced from Artificial Analysis); "similar intelligence" means within 4 points on its Intelligence Index. Developer locations are where each company is based, not where a model is served. Groups under 3 models are not ranked.

Measured cost per task

API cost recorded in our benchmark runs, excluding judging.

TaskBenchmarkCost per task
Product listingCatalogBench≈ $0.000
DeckDeckBench≈ $0.009
Game of blitz chessBulletBench≈ $0.007

Cheapest model that passes

On our benchmarks that define a pass for each task: what one passing answer cost in our run, failed attempts included, and whether a cheaper model passed.

  • CatalogBench · cost per 1,000 accepted listings
    GPT-6 Luna is the cheapest of the 23 models we could cost this way: $0.70 (114 of 168 attempts).
  • DeckBench · cost per presentable deck
    GPT-6 Luna: no presentable decks (0 of 6 tasks). Cheapest: GPT-6.1 Terra at $0.13 (3 of 6 tasks).

Cost per attempt divided by the share that passed, at the run date's prices, excluding judging. Each model at its cheapest setting per passing answer. With a handful of tasks, a pass or two either way is within chance. No estimate where a run had missing outputs, because their cost may not have been recorded.

How good is it, and for what?

Each use case ranks the available models its benchmarks measured, then those priced under $0.25 per million tokens. The bar shows its percentile in that field, best to the right. Open a row for the results behind it.

Product listingsTurning a sparse product feed and photos into listings that can go live 6th of 27 available · 1st of 3 at its price
BenchmarkGPT-6 LunaRank among availableBest available
CatalogBenchCatalogBench: reliably publish-ready 55.4% 8th of 27 GPT-6 Astra 75.0%
CatalogBenchCatalogBench (sales brief): reliably publish-ready 51.8% 5th of 27 GPT-6 Astra 69.6%
Decks from an analysisTurning a finished analysis into a deck you could present as it is 9th of 28 available · 1st of 3 at its price
BenchmarkGPT-6 LunaRank among availableBest available
DeckBenchDeckBench: deck rating 1,296 9th of 28 GPT-6 Astra 1,606
User surveysPlanning a user survey and reading its results without being misled Not measured
  • SurveyBench: has not measured this model.
Marketing planningPlanning a year of ad spend without overspending 5th of 28 available · 1st of 3 at its price
BenchmarkGPT-6 LunaRank among availableBest available
ROASBenchROASBench: overall score 53.4 5th of 28 GPT-6.1 Terra 58.4
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 50th of 160 available · 5th of 38 at its price
BenchmarkGPT-6 LunaRank among availableBest available
Artificial AnalysisArtificial Analysis: Terminal-Bench 4.0 · reported 12.6% 33rd of 106 Claude Sonnet 5.5 63.6%
  • tau2-bench: pending. Its latest results are dated 26 May 2026, before this model was released.
  • Berkeley Function Calling Leaderboard (BFCL) V4: pending. Its latest results are dated 16 Dec 2025, before this model was released.
  • OpenHands Index: pending. Its latest results are dated 30 Jun 2026, before this model was released.
  • Microsoft STATE-Bench: pending. Its latest results are dated 29 May 2026, before this model was released.
  • Vending-Bench 2: has not measured this model.
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 25th of 51 available · 1st of 4 at its price
BenchmarkGPT-6 LunaRank among availableBest available
APEX-AgentsAPEX-Agents: tasks passed · reported 44.3% 27th of 37 Claude Sonnet 5.5 75.5%
GDP.pdfGDP.pdf: professional document tasks · reported 23.0% 14th of 31 GPT-5.6 Sol 30.7%
  • Remote Labor Index: has not measured this model.
Reasoning and knowledgeHard questions across science, maths and general knowledge 46th of 188 available · 4th of 42 at its price
BenchmarkGPT-6 LunaRank among availableBest available
Artificial AnalysisArtificial Analysis: Artificial Analysis Intelligence Index · reported 38.1 37th of 188 Claude Opus 5.5 57.6
Artificial AnalysisArtificial Analysis: Humanity's Last Exam · reported 38.5% 48th of 187 Claude Opus 5.5 61.4%
WritingWhat people prefer in blind comparisons, and judged writing quality 61st of 173 available · 3rd of 40 at its price
BenchmarkGPT-6 LunaRank among availableBest available
Arena (formerly LMArena)Arena Text: overall · reported 1,443 58th of 156 Gemini 4 Argon 1,525
  • UGI Leaderboard: has not measured this model.
Sticking to the factsSummarising without inventing things, and factual answers 89th of 143 available · 16th of 31 at its price
BenchmarkGPT-6 LunaRank among availableBest available
Arena (formerly LMArena)Arena Text factuality: overall · reported 1,430 76th of 112 Gemini 4 Argon 1,511
SimpleQA Verified (Epoch AI)SimpleQA Verified: correct answers · reported 41.4% 37th of 68 GPT-6 Astra 75.6%
  • Vectara Hallucination Leaderboard: has not measured this model.
Speed under pressureGood decisions against a real clock (fast chess) 21st of 31 available · 3rd of 3 at its price
BenchmarkGPT-6 LunaRank among availableBest available
BulletBenchBulletBench Bullet 60s: ladder Elo 249 20th of 29 Gemini 3.8 Flash 1,068

Nearest alternatives

Against GPT-6 Luna at $0.20 blended, Intelligence Index 38.1. "Similar intelligence" means within 4 points or better.

Where to run it

6 hosts on OpenRouter at the standard service tier, with a median of $0.21 blended per million tokens. Cheapest first.

HostQuantisationInputOutputBlended
Amazon Bedrocknot stated$0.10$0.50$0.20
Azurenot stated$0.10$0.50$0.20
OpenAInot stated$0.10$0.50$0.20
Amazon Bedrocknot stated$0.11$0.55$0.22
Azurenot stated$0.11$0.55$0.22
Azurenot stated$0.11$0.55$0.22

Other service tiers, not counted above: OpenAI fast $0.20 in, $1.00 out; OpenAI flex $0.05 in, $0.25 out. Flex and batch are cheaper but slower or queued; priority and fast cost more.

Prices per million tokens from OpenRouter's public endpoint list, checked 5 Nov 2026.

Against the alternatives

The models you would most likely weigh it against: the leaders of the groups above.

Scroll sideways to see every alternative.

GPT-6 LunaClaude Haiku 5.5
most intelligent at under $0.25 per million tokens
MiMo-V2.6-Flash
cheapest with similar intelligence
DeepSeek-V4.1-Flash
fastest with similar intelligence
Claude Opus 5.5
most intelligent available today
Price per million tokens $0.20 $0.20$0.18$0.52$8.00
Tokens a second 127 2404636996
Intelligence Index 38.1 43.437.939.557.6
Spring Prompt overall 39 44–5384
Rank among available models, by use case
Product listings 6th of 27 15th–11th10th
Decks from an analysis 9th of 28 14th––4th
User surveys – 4th––3rd
Marketing planning 5th of 28 26th––7th
Agents and tool use 50th of 160 23rd35th28th2nd
Professional work 25th of 51 –––3rd
Reasoning and knowledge 46th of 188 22nd53rd40th1st
Writing 61st of 173 –56th29th5th
Sticking to the facts 89th of 143 –45th23rd3rd
Speed under pressure 21st of 31 19th–––

Shaded figures are better than GPT-6 Luna's. A dash means no result.

Research and examples

In the playground: our pick in 0 of 20 tests, 84% of checks passed →

Where it has been measured

3rd most intelligent of the 42 models priced under $0.25 per million tokens, and 3rd cheapest of the 21 models with similar intelligence. Among models available today, in the top quarter (by percentile) for marketing planning, reasoning and knowledge, and product listings.

Every result

121 published results from 9 sources, each in the source's own units, with the configuration that produced it.

BulletBench · measured by Spring Prompt · 10 results
MeasureValueRankConfigurationDated
BulletBench Blitz 3+2: cost per gameUS dollars, lower is better $0.0072 1 of 17 GPT-6 Luna
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: games lost on time% of games, lower is better 50.0% 14 of 17 GPT-6 Luna
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: invalid moves% of moves, lower is better 0.0% 1 of 17 GPT-6 Luna
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: ladder Eloladder Elo, higher is better 557
370–727
9 of 17 GPT-6 Luna
provider default reasoning
29 Sep 2026
BulletBench Blitz 3+2: median move timemilliseconds, lower is better 7.0 s 14 of 17 GPT-6 Luna
provider default reasoning
29 Sep 2026
BulletBench Bullet 60s: cost per gameUS dollars, lower is better $0.0016 6 of 44 GPT-6 Luna (no reasoning)
none reasoning
1 Oct 2026
BulletBench Bullet 60s: games lost on time% of games, lower is better 50.0% 23 of 44 GPT-6 Luna (no reasoning)
none reasoning
1 Oct 2026
BulletBench Bullet 60s: invalid moves% of moves, lower is better 0.0% 1 of 44 GPT-6 Luna (no reasoning)
none reasoning
1 Oct 2026
BulletBench Bullet 60s: ladder Eloladder Elo, higher is better 249
0–453
31 of 44 GPT-6 Luna (no reasoning)
none reasoning
1 Oct 2026
BulletBench Bullet 60s: median move timemilliseconds, lower is better 1.5 s 22 of 44 GPT-6 Luna (no reasoning)
none reasoning
1 Oct 2026

Not shown: Not your cost: prices are those charged on the run date.

CatalogBench · measured by Spring Prompt · 31 results
MeasureValueRankConfigurationDated
CatalogBench (sales brief): channel rules broken% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): claims to check% of products, lower is better 0.0% 1 of 27 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): content quality% of checks, higher is better 97.4% 11 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): failed outputs% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): missing UK information% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): not findable% of products, lower is better 10.1% 26 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): publish-ready listings% of products, higher is better 67.9% 6 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): reliably publish-ready% of products, higher is better 51.8% 5 of 29 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claims% of products, lower is better 8.3% 6 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): unsupported claimsclaims per product, lower is better 0.35 6 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong attributes% of products, lower is better 19.1% 16 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench (sales brief): wrong category or variant% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: channel compliance% of products, higher is better 100.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: channel rules broken% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: claims to check% of products, lower is better 0.0% 1 of 27 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: conflicts caught% of conflicts, higher is better 100.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: content quality% of checks, higher is better 97.0% 14 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: cost per productUS dollars, lower is better $0.0005 2 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: decision accuracy% of decisions, higher is better 97.0% 11 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: failed outputs% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: field accuracy% of missing fields, higher is better 93.7% 13 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: invented values% of filled values, lower is better 1.7% 14 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: missing UK information% of products, lower is better 0.0% 1 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: not findable% of products, lower is better 10.1% 26 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: publish-ready listings% of products, higher is better 67.9% 10 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: reliably publish-ready% of products, higher is better 55.4% 8 of 29 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: time per productseconds, lower is better 8.20 6 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claims% of products, lower is better 8.9% 9 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: unsupported claimsclaims per product, lower is better 0.31 7 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: wrong attributes% of products, lower is better 18.5% 17 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026
CatalogBench: wrong category or variant% of products, lower is better 1.8% 5 of 30 GPT-6 Luna
provider default reasoning
30 Sep 2026

Not shown: Not your catalogue: invented products with generated images, so a model's result on your own feed can differ.

DeckBench · measured by Spring Prompt · 17 results
MeasureValueRankConfigurationDated
DeckBench: accurate decks% of tasks, higher is better 50.0% 12 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: caveat dropped% of tasks, lower is better 0.0% 1 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: clean layout% of tasks, higher is better 66.7% 7 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: cost per deckUS dollars, lower is better $0.0087 3 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: deck ratingrating, higher is better 1,296
1,233–1,382
9 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: design quality% of the maximum, higher is better 78.1% 7 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: draft figure quoted% of tasks, lower is better 33.3% 15 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: findings missing% of tasks, lower is better 0.0% 1 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: head-to-head win rate% of comparisons, higher is better 72.5% 8 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: layout defects% of tasks, lower is better 33.3% 7 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: misleading metric used% of tasks, lower is better 0.0% 1 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: presentable decks% of tasks, higher is better 0.0% 13 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: recommendation late or wrong% of tasks, lower is better 16.7% 12 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: slides needing work% of tasks, lower is better 16.7% 3 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: time per deckseconds, lower is better 101 14 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: unsupported claims% of tasks, lower is better 0.0% 1 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026
DeckBench: unsupported numbers% of tasks, lower is better 0.0% 1 of 29 GPT-6 Luna
provider default reasoning
6 Oct 2026

Not shown: Not your analysis or your brand: invented companies, six public tasks, one deck per model per task at the provider's default reasoning setting.

ROASBench · measured by Spring Prompt · 9 results
MeasureValueRankConfigurationDated
ROASBench: audience scorescore out of 100, higher is better 68.3 4 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: business scorescore out of 100, higher is better 54.8 5 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: consistency scorescore out of 100, higher is better 35.8 6 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: contribution profitsimulated US dollars, higher is better $738,879 5 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: cost of a runUS dollars, lower is better $0.0142 2 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: months over budgetmonths of 12, lower is better 0 1 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: overall scorescore out of 100, higher is better 53.4 5 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: planning scorescore out of 100, higher is better 54.8 15 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026
ROASBench: return on ad spendprofit per $1 spent, higher is better $0.66 6 of 30 GPT-6 Luna
provider default reasoning
29 Sep 2026

Not shown: Not real ad performance: the market is a deterministic simulation of one invented skincare brand.

APEX-Agents · reported by APEX-Agents · 1 result
MeasureValueRankConfigurationDated
APEX-Agents: tasks passed% of tasks, higher is better 44.3% 29 of 39 GPT-6 Luna (max reasoning) 1 Oct 2026

Not shown: Not your firm's documents or tools; graded by rubric, not by a client.

Arena (formerly LMArena) · reported by Arena (formerly LMArena) · 11 results
MeasureValueRankConfigurationDated
Arena Agent: confirmed task successIPS effect estimate, higher is better -0.00
-0.03–0.02
30 of 50 GPT-6 Luna (max reasoning, arena agent) 2 Oct 2026
Arena Agent: praise over complaintIPS effect estimate, higher is better -0.02
-0.05–0.02
32 of 50 GPT-6 Luna (max reasoning, arena agent) 2 Oct 2026
Arena Agent: steerabilityIPS effect estimate, higher is better 0.04
0.03–0.05
17 of 50 GPT-6 Luna (max reasoning, arena agent) 2 Oct 2026
Arena Agent: tool groundingIPS effect estimate, higher is better -0.00
-0.00–-0.00
41 of 50 GPT-6 Luna (max reasoning, arena agent) 2 Oct 2026
Arena Text factuality: overallArena rating, higher is better 1,430
1,424–1,436
115 of 182 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: business, management and financeArena rating, higher is better 1,447
1,432–1,461
82 of 406 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: creative writingArena rating, higher is better 1,413
1,400–1,427
90 of 411 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: expert promptsArena rating, higher is better 1,498
1,480–1,516
46 of 363 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: instruction followingArena rating, higher is better 1,448
1,438–1,458
68 of 413 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: overallArena rating, higher is better 1,443
1,436–1,449
90 of 413 GPT-6 Luna (max reasoning) 2 Oct 2026
Arena Text: writing, literature and languageArena rating, higher is better 1,421
1,409–1,433
92 of 412 GPT-6 Luna (max reasoning) 2 Oct 2026

Not shown: Not a success rate: the estimated effect of choosing this model as the agent's orchestrator, against an even mix of the models tested, inside Arena's own harness.

Artificial Analysis · reported by Artificial Analysis · 40 results
MeasureValueRankConfigurationDated
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 32.9 103 of 413 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 21.5 193 of 413 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 38.1 76 of 413 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 29.9 119 of 413 GPT-6 Luna (medium reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 18.5 223 of 413 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Artificial Analysis Intelligence Indexindex score, higher is better 34.6 87 of 413 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 32.9% 131 of 411 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 20.3% 192 of 411 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 38.5% 94 of 411 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 28.3% 153 of 411 GPT-6 Luna (medium reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 8.6% 287 of 411 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Humanity's Last Exam% of questions, higher is better 34.3% 122 of 411 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 79.3% 90 of 399 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 74.0% 157 of 399 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 83.3% 18 of 399 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 78.3% 103 of 399 GPT-6 Luna (medium reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 39.7% 311 of 399 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Long-context reasoning (AA-LCR)% of questions, higher is better 80.0% 71 of 399 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 122 53 of 137 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 115 58 of 137 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 127 52 of 137 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 113 60 of 137 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Output speedtokens per second, higher is better 112 61 of 137 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 50.3% 103 of 197 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 46.9% 130 of 197 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 54.6% 58 of 197 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 50.9% 97 of 197 GPT-6 Luna (medium reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 43.1% 145 of 197 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: SciCode% of problems, higher is better 51.7% 87 of 197 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 4.5% 102 of 193 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 0.0% 141 of 193 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 12.6% 76 of 193 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 2.5% 109 of 193 GPT-6 Luna (medium reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 1.5% 118 of 193 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Terminal-Bench 4.0% of tasks, higher is better 8.1% 90 of 193 GPT-6 Luna (xhigh reasoning) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 17.1 80 of 137 GPT-6 Luna (high reasoning) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 3.52 39 of 137 GPT-6 Luna (low reasoning) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 116 128 of 137 GPT-6 Luna (max reasoning) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 0.74 20 of 137 GPT-6 Luna (no reasoning) 20 Oct 2026
Artificial Analysis: Time to first answer tokenseconds, lower is better 37.7 103 of 137 GPT-6 Luna (xhigh reasoning) 20 Oct 2026

Not shown: Not business work, and a blend: read the parts for any one task.

GDP.pdf · reported by GDP.pdf · 1 result
MeasureValueRankConfigurationDated
GDP.pdf: professional document tasks% of rubric, higher is better 23.0% 20 of 49 GPT-6 Luna (max reasoning) 1 Oct 2026

Not shown: Not your documents; 100 tasks, so small differences are noise.

SimpleQA Verified (Epoch AI) · reported by SimpleQA Verified (Epoch AI) · 1 result
MeasureValueRankConfigurationDated
SimpleQA Verified: correct answers% of questions, higher is better 41.4% 41 of 85 GPT-6 Luna (max reasoning) 22 Sep 2026

Not shown: Not answers grounded in your documents; tests what the model remembers.

Embed a score badge

A badge for each of our benchmarks that ranked GPT-6 Luna, from the current results. Paste the code into a page or README; it links back to the benchmark.

GPT-6 Luna is #8 of 27 on CatalogBench (Reliably publish-ready, factual brief: 55.4%), measured by Spring Prompt CatalogBench · Reliably publish-ready, factual brief
<a href="https://springprompt.com/benchmarks/catalogbench?focus=openai-gpt-6-luna"><img src="https://springprompt.com/badge/catalogbench/openai-gpt-6-luna.svg" alt="GPT-6 Luna is #8 of 27 on CatalogBench (Reliably publish-ready, factual brief: 55.4%), measured by Spring Prompt" height="20"></a>
GPT-6 Luna is #9 of 28 on DeckBench (Deck rating: 1,296), measured by Spring Prompt DeckBench · Deck rating
<a href="https://springprompt.com/benchmarks/deckbench?focus=openai-gpt-6-luna"><img src="https://springprompt.com/badge/deckbench/openai-gpt-6-luna.svg" alt="GPT-6 Luna is #9 of 28 on DeckBench (Deck rating: 1,296), measured by Spring Prompt" height="20"></a>
GPT-6 Luna is #5 of 28 on ROASBench (Score: 53.4), measured by Spring Prompt ROASBench · Score
<a href="https://springprompt.com/benchmarks/roasbench?focus=openai-gpt-6-luna"><img src="https://springprompt.com/badge/roasbench/openai-gpt-6-luna.svg" alt="GPT-6 Luna is #5 of 28 on ROASBench (Score: 53.4), measured by Spring Prompt" height="20"></a>
GPT-6 Luna is #27 of 36 on BulletBench (Ladder Elo: 249), measured by Spring Prompt BulletBench · Ladder Elo
<a href="https://springprompt.com/benchmarks/bulletbench?focus=openai-gpt-6-luna"><img src="https://springprompt.com/badge/bulletbench/openai-gpt-6-luna.svg" alt="GPT-6 Luna is #27 of 36 on BulletBench (Ladder Elo: 249), measured by Spring Prompt" height="20"></a>

Test GPT-6 Luna on your own prompt.

Benchmarks aren't your task. Run it next to cheaper and newer models on your prompt, with every answer checked and costed.