Compare / GPT-6.1 Terra vs GPT-6 Luna

GPT-6.1 Terra vs GPT-6 Luna

Which is cheaper, which is more capable, and which does better at each kind of work, from 92 results on 6 sources that measured both models.

OpenAIGPT-6.1 Terra Released 26 Oct 2026 · $1.00 in · $5.00 out
OpenAIGPT-6 Luna Released 22 Sep 2026 · $0.10 in · $0.50 out
GPT-6.1 Terra wins 4229 too close to callGPT-6 Luna wins 21

Of the 92 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 19 November 2026 (catalogue release 2026-11-19-3608133e8331).

GPT-6.1 Terra or GPT-6 Luna? The short answer

Pick GPT-6.1 Terra if quality matters most: it scores higher (Spring Prompt overall 77 against 38). Pick GPT-6 Luna if cost matters more: it costs 7.5 times less per product listing on CatalogBench.

  • Spring Prompt overall: GPT-6.1 Terra 77, GPT-6 Luna 38 (out of 100; how the score works).
  • GPT-6 Luna is cheaper to run: $0.000 against $0.004 per product listing on CatalogBench (7.5 times less).
  • GPT-6.1 Terra is newer: released 26 Oct 2026, against 22 Sep 2026 for GPT-6 Luna.
  • On our own benchmarks, on the results both models have: GPT-6.1 Terra does better for product listings, user surveys and marketing planning.
  • On third-party benchmarks, on the results both models have: GPT-6.1 Terra does better for agents and tool use and reasoning and knowledge.

Head to head

77
Spring Prompt overall
38
$2.00
Price per million tokens, blendedlower is better
$0.20
1,050,000
Context window, tokens
1,050,000

Every fact side by side

GPT-6.1 TerraGPT-6 Luna
Price per million tokens$1.00 in · $5.00 out$0.10 in · $0.50 out
Blended (3 in : 1 out)$2.00$0.20
Cheapest host, blended$2.00 (Azure)$0.20 (Amazon Bedrock)
Cost per product listing (CatalogBench)$0.004$0.000
Cached input, per million$0.050 cached input$0.010 cached input
Intelligence Index–38.1
Spring Prompt overall7738
Output speed, tokens a second–127
Context window1,050,000 tokens1,050,000 tokens
Released26 Oct 2026 (newer)22 Sep 2026
WeightsClosedClosed
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live 6 of 35 10 of 35
Decks from an analysisTurning a finished analysis into a deck you could present as it is 9 of 35 12 of 35
User surveysPlanning a user survey and reading its results without being misled 10 of 26 14 of 26
Marketing planningPlanning a year of ad spend without overspending 2 of 35 7 of 35
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 7 of 159 53 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects not measured 25 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 22 of 187 48 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality not measured 61 of 169
Sticking to the factsSummarising without inventing things, and factual answers not measured 89 of 139
Speed under pressureGood decisions against a real clock (fast chess) 21 of 38 28 of 38

Their slides, side by side

The same brief for both: Manchester store investment review for Hearthside Coffee, an invented company. First slide of each deck from DeckBench.

First slide of GPT-6.1 Terra's deck for Hearthside Coffee
GPT-6.1 Terra · deck rating 1,280 · needs work
First slide of GPT-6 Luna's deck for Hearthside Coffee
GPT-6 Luna · deck rating 1,252 · needs work

Every slide, side by side →

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureGPT-6.1 TerraGPT-6 Luna
BulletBench measured by Spring Prompt · 3 results
Cost per game · Blitz 3+2US dollars, lower is better 9 · 1 of 23 measured $0.0706 $0.0072
Cost per game · Bullet 60sUS dollars, lower is better 47 · 21 of 62 measured $0.0172low reasoning $0.0016no reasoning
Ladder Elo · Bullet 60sladder Elo, higher is better 46 · 52 of 62 measured 358low reasoning 249no reasoning
CatalogBench measured by Spring Prompt · 3 results
Cost per productUS dollars, lower is better 9 · 2 of 36 measured $0.0036 $0.0005
Reliably publish-ready% of products, higher is better 9 · 13 of 35 measured 64.3% 55.4%
Reliably publish-ready · sales brief% of products, higher is better 4 · 7 of 35 measured 67.9% 51.8%
DeckBench measured by Spring Prompt · 3 results
Cost per deckUS dollars, lower is better 16 · 3 of 36 measured $0.0672 $0.0087
Deck ratingrating, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 9 · 12 of 36 measured 1,280 1,252
Accurate decks% of tasks, higher is better 1 · 17 of 36 measured 100.0% 50.0%
ROASBench measured by Spring Prompt · 3 results
Overall scorescore out of 100, higher is better 2 · 7 of 36 measured 58.4 53.4
Audience scorescore out of 100, higher is better 1 · 7 of 36 measured 69.4 68.3
Business scorescore out of 100, higher is better 3 · 7 of 36 measured 59.1 54.8
SurveyBench measured by Spring Prompt · 3 results
Analysis ratingrating, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 10 · 14 of 27 measured 1,217 1,072
Cost per taskUS dollars, lower is better 9 · 1 of 27 measured $0.0569 $0.0088
Analysis scorepoints of 100, higher is better 6 · 7 of 27 measured 91.6 90.8
Artificial Analysis reported by Artificial Analysis · 3 results
Humanity's Last Exam% of questions, higher is better 21 · 53 of 250 measured 46.2%max reasoning 38.5%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 7 · 37 of 120 measured 55.6%max reasoning 12.6%max reasoning
Long-context reasoning (AA-LCR)% of questions, higher is better 20 · 13 of 241 measured 82.7%max reasoning 83.3%max reasoning
Show all 92 results (74 more)
MeasureGPT-6.1 TerraGPT-6 Luna
BulletBench measured by Spring Prompt · 7 results
Games lost on time · Blitz 3+2% of games, lower is better 10 · 15 of 23 measured 25.0% 50.0%
Games lost on time · Bullet 60s% of games, lower is better 52 · 39 of 62 measured 75.0%low reasoning 50.0%no reasoning
Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 23 measured 0.0% 0.0%
Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 62 measured 0.0%low reasoning 0.0%no reasoning
Ladder Elo · Blitz 3+2ladder Elo, higher is better 12 · 14 of 23 measured 683 557
Median move time · Blitz 3+2milliseconds, lower is better 8 · 19 of 23 measured 3.7 s 7.0 s
Median move time · Bullet 60smilliseconds, lower is better 53 · 39 of 62 measured 2.8 slow reasoning 1.5 sno reasoning
CatalogBench measured by Spring Prompt · 28 results
Channel compliance% of products, higher is better 35 · 1 of 36 measured 98.2% 100.0%
Channel rules broken% of products, lower is better 33 · 1 of 36 measured 1.8% 0.0%
Channel rules broken · sales brief% of products, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Claims to check% of products, lower is better 1 · 1 of 34 measured 0.0% 0.0%
Claims to check · sales brief% of products, lower is better 1 · 1 of 34 measured 0.0% 0.0%
Conflicts caught% of conflicts, higher is better 1 · 1 of 36 measured 100.0% 100.0%
Content quality% of checks, higher is better 13 · 20 of 36 measured 97.7% 97.0%
Content quality · sales brief% of checks, higher is better 10 · 18 of 36 measured 98.3% 97.4%
Decision accuracy% of decisions, higher is better 8 · 16 of 36 measured 98.1% 97.0%
Failed outputs% of products, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Failed outputs · sales brief% of products, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Field accuracy% of missing fields, higher is better 10 · 18 of 36 measured 94.4% 93.7%
Invented values% of filled values, lower is better 16 · 18 of 36 measured 1.5% 1.7%
Missing UK information% of products, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Missing UK information · sales brief% of products, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Not findable% of products, lower is better 33 · 32 of 36 measured 10.7% 10.1%
Not findable · sales brief% of products, lower is better 35 · 32 of 36 measured 13.7% 10.1%
Publish-ready listings% of products, higher is better 10 · 17 of 36 measured 74.4% 67.9%
Publish-ready listings · sales brief% of products, higher is better 5 · 8 of 36 measured 74.4% 67.9%
Time per productseconds, lower is better 12 · 8 of 36 measured 10.1 8.20
Unsupported claimsclaims per product, lower is better 6 · 13 of 36 measured 0.03 0.31
Unsupported claims% of products, lower is better 3 · 13 of 36 measured 0.6% 8.9%
Unsupported claims · sales briefclaims per product, lower is better 5 · 8 of 36 measured 0.09 0.35
Unsupported claims · sales brief% of products, lower is better 2 · 8 of 36 measured 1.8% 8.3%
Wrong attributes% of products, lower is better 12 · 24 of 36 measured 14.9% 18.5%
Wrong attributes · sales brief% of products, lower is better 11 · 23 of 36 measured 14.9% 19.1%
Wrong category or variant% of products, lower is better 7 · 7 of 36 measured 1.8% 1.8%
Wrong category or variant · sales brief% of products, lower is better 7 · 1 of 36 measured 1.8% 0.0%
DeckBench measured by Spring Prompt · 14 results
Caveat dropped% of tasks, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Clean layout% of tasks, higher is better 5 · 10 of 36 measured 83.3% 66.7%
Design quality% of the maximum, higher is better 8 · 10 of 36 measured 80.8% 78.1%
Draft figure quoted% of tasks, lower is better 1 · 21 of 36 measured 0.0% 33.3%
Findings missing% of tasks, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other deck was; the win rate counts wins against whichever decks a model happened to meet, so the two can disagree. 26 · 9 of 36 measured 41.9% 70.9%
Layout defects% of tasks, lower is better 5 · 10 of 36 measured 16.7% 33.3%
Misleading metric used% of tasks, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Presentable decks% of tasks, higher is better 5 · 17 of 36 measured 50.0% 0.0%
Recommendation late or wrong% of tasks, lower is better 1 · 17 of 36 measured 0.0% 16.7%
Slides needing work% of tasks, lower is better 12 · 5 of 36 measured 33.3% 16.7%
Time per deckseconds, lower is better 19 · 17 of 36 measured 135 101
Unsupported claims% of tasks, lower is better 1 · 1 of 36 measured 0.0% 0.0%
Unsupported numbers% of tasks, lower is better 1 · 1 of 36 measured 0.0% 0.0%
ROASBench measured by Spring Prompt · 6 results
Consistency scorescore out of 100, higher is better 1 · 10 of 36 measured 48.8 35.8
Contribution profitsimulated US dollars, higher is better 10 · 9 of 36 measured $701,059 $738,879
Cost of a runUS dollars, lower is better 11 · 2 of 36 measured $0.11 $0.0142
Months over budgetmonths of 12, lower is better 1 · 1 of 36 measured 0 0
Planning scorescore out of 100, higher is better 10 · 20 of 36 measured 55.0 54.8
Return on ad spendprofit per $1 spent, higher is better 3 · 10 of 36 measured $0.71 $0.66
SurveyBench measured by Spring Prompt · 18 results
Head-to-head win rate% of comparisons, higher is betterThe rating weighs each win by how strong the other analysis was; the win rate counts wins against whichever analyses a model happened to meet, so the two can disagree. 11 · 17 of 27 measured 56.3% 47.1%
Leading or double-barrelled questions% of tasks, lower is better 5 · 5 of 27 measured 11.1% 11.1%
Mistaken next steps% of tasks, lower is better 5 · 5 of 27 measured 11.1% 11.1%
Numbers right% of answers, higher is better 1 · 21 of 27 measured 100.0% 98.0%
Objectives unanswerable% of tasks, lower is better 18 · 1 of 27 measured 11.1% 0.0%
Output tokens per tasktokens, lower is better 2 · 4 of 17 measured 8,112 14,333
Precision ignored% of tasks, lower is better 1 · 1 of 27 measured 0.0% 0.0%
Research scorepoints of 100, higher is better 6 · 7 of 27 measured 92.2 91.7
Sound research% of tasks, higher is better 6 · 6 of 27 measured 33.3% 33.3%
Stakeholder's view presumed% of tasks, lower is better 12 · 12 of 27 measured 11.1% 11.1%
Survey plan scorepoints of 100, higher is better 10 · 9 of 27 measured 93.6 94.0
Survey structure% of tasks, lower is better 12 · 12 of 27 measured 22.2% 22.2%
Time per taskseconds, lower is better 7 · 12 of 27 measured 92.1 138
Traps handled% of traps, higher is better 11 · 11 of 27 measured 94.2% 94.2%
Traps missed% of tasks, lower is better 8 · 8 of 27 measured 33.3% 33.3%
Unsupported findings% of tasks, lower is better 1 · 7 of 27 measured 0.0% 22.2%
Wrong numbers% of tasks, lower is better 1 · 14 of 27 measured 0.0% 11.1%
Wrong recommendation% of tasks, lower is better 5 · 5 of 27 measured 11.1% 11.1%
Artificial Analysis reported by Artificial Analysis · 1 result
SciCode% of problems, higher is better 27 · 31 of 121 measured 55.1%max reasoning 54.6%max reasoning

Quick answers

Is GPT-6.1 Terra better than GPT-6 Luna?

Pick GPT-6.1 Terra if quality matters most: it scores higher (Spring Prompt overall 77 against 38). Pick GPT-6 Luna if cost matters more: it costs 7.5 times less per product listing on CatalogBench.

Which is cheaper, GPT-6.1 Terra or GPT-6 Luna?

GPT-6 Luna is cheaper to run: $0.000 against $0.004 per product listing on CatalogBench (7.5 times less).

Which is better for product listings, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra ranks 6 of 35 models on sale for turning a sparse product feed and photos into listings that can go live, against 10 for GPT-6 Luna.

Which is better for decks from an analysis, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra ranks 9 of 35 models on sale for turning a finished analysis into a deck you could present as it is, against 12 for GPT-6 Luna.

Which is better for user surveys, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra ranks 10 of 26 models on sale for planning a user survey and reading its results without being misled, against 14 for GPT-6 Luna.

Which is better for marketing planning, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra ranks 2 of 35 models on sale for planning a year of ad spend without overspending, against 7 for GPT-6 Luna.

Which is better for agents and tool use, GPT-6.1 Terra or GPT-6 Luna?

GPT-6.1 Terra ranks 7 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 53 for GPT-6 Luna.

Run GPT-6.1 Terra and GPT-6 Luna on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.