Compare / GPT-6.1 Sol vs GPT-6.1 Terra
GPT-6.1 Sol vs GPT-6.1 Terra
Which is cheaper, which is more capable, and which does better at each kind of work, from 83 results on 4 sources that measured both models.
OpenAIGPT-6.1 Sol
OpenAIGPT-6.1 Terra
GPT-6.1 Sol wins 3136 too close to callGPT-6.1 Terra wins 16
Of the 83 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.
Results as of 4 November 2026 (catalogue release 2026-11-04-56cfea57899a).
GPT-6.1 Sol or GPT-6.1 Terra? The short answer
They score about the same, so pick GPT-6.1 Terra: it costs 2.4 times less per product listing on CatalogBench.
- Spring Prompt overall: GPT-6.1 Sol 80, GPT-6.1 Terra 81 (out of 100; how the score works).
- GPT-6.1 Terra is cheaper to run: $0.004 against $0.008 per product listing on CatalogBench (2.4 times less).
- On our own benchmarks, on the results both models have: GPT-6.1 Sol does better for product listings, user surveys and speed under pressure.
Head to head
80
Spring Prompt overall
81
$4.00
Price per million tokens, blendedlower is better
$2.20
1,050,000
Context window, tokens
1,050,000
Every fact side by side
| GPT-6.1 Sol | GPT-6.1 Terra | |
|---|---|---|
| Price per million tokens | $2.00 in · $10.00 out | $1.10 in · $5.50 out |
| Blended (3 in : 1 out) | $4.00 | $2.20 |
| Cheapest host, blended | $4.00 (Amazon Bedrock) | $2.00 (Azure) |
| Cost per product listing (CatalogBench) | $0.008 | $0.004 |
| Cached input, per million | $0.10 cached input | – |
| Intelligence Index | 51.8 | – |
| Spring Prompt overall | 80 | 81 |
| Output speed, tokens a second | 53 | – |
| Context window | 1,050,000 tokens | 1,050,000 tokens |
| Released | 29 Sep 2026 | – |
| Weights | Closed | Closed |
| Developer based in | the US | the 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
2 of 25
4 of 25
Decks from an analysisTurning a finished analysis into a deck you could present as it is
3 of 26
7 of 26
User surveysPlanning a user survey and reading its results without being misled
3 of 15
5 of 15
Marketing planningPlanning a year of ad spend without overspending
not measured
1 of 26
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls
5 of 160
not measured
Professional workReal tasks from banking, consulting and law, business documents and freelance projects
9 of 51
not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge
6 of 188
not measured
WritingWhat people prefer in blind comparisons, and judged writing quality
16 of 173
not measured
Sticking to the factsSummarising without inventing things, and factual answers
18 of 143
not measured
Speed under pressureGood decisions against a real clock (fast chess)
3 of 29
16 of 29
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.
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.
| Measure | GPT-6.1 Sol | GPT-6.1 Terra |
|---|---|---|
| BulletBench measured by Spring Prompt · 3 results | ||
| Cost per game · Blitz 3+2US dollars, lower is better 8 · 7 of 16 measured | $0.0767 | $0.0706 |
| Cost per game · Bullet 60sUS dollars, lower is better 27 · 24 of 34 measured | $0.0210 | $0.0172low reasoning |
| Ladder Elo · Bullet 60sladder Elo, higher is better 2 · 23 of 34 measured | 928Ultrafast (low reasoning) | 358low reasoning |
| CatalogBench measured by Spring Prompt · 3 results | ||
| Cost per productUS dollars, lower is better 12 · 7 of 26 measured | $0.0084 | $0.0036 |
| Reliably publish-ready% of products, higher is better 1 · 5 of 25 measured | 75.0% | 64.3% |
| Reliably publish-ready · sales brief% of products, higher is better 2 · 2 of 25 measured | 67.9% | 67.9% |
| DeckBench measured by Spring Prompt · 3 results | ||
| Cost per deckUS dollars, lower is better 18 · 13 of 26 measured | $0.15 | $0.0672 |
| 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. 3 · 7 of 26 measured | 1,446 | 1,355 |
| Accurate decks% of tasks, higher is better 1 · 1 of 26 measured | 100.0% | 100.0% |
| 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. 3 · 5 of 15 measured | 1,482 | 1,246 |
| Cost per taskUS dollars, lower is better 10 · 5 of 14 measured | $0.14 | $0.0569 |
| Analysis scorepoints of 100, higher is better 1 · 4 of 15 measured | 96.6 | 91.6 |
Show all 83 results (71 more)
| Measure | GPT-6.1 Sol | GPT-6.1 Terra |
|---|---|---|
| BulletBench measured by Spring Prompt · 12 results | ||
| Ladder Elo · Lightning 10+1ladder Elo, higher is better 5 · 17 of 25 measured | 584Ultrafast (low reasoning) | 491low reasoning |
| Cost per game · Lightning 10+1US dollars, lower is better 25 · 15 of 25 measured | $0.16Ultrafast | $0.0064low reasoning |
| Games lost on time · Blitz 3+2% of games, lower is better 1 · 8 of 16 measured | 0.0%Ultrafast | 25.0% |
| Games lost on time · Bullet 60s% of games, lower is better 14 · 27 of 34 measured | 25.0%Ultrafast (low reasoning) | 75.0%low reasoning |
| Games lost on time · Lightning 10+1% of games, lower is better 17 · 19 of 25 measured | 50.0%Ultrafast (low reasoning) | 62.5%low reasoning |
| Invalid moves · Blitz 3+2% of moves, lower is better 1 · 1 of 16 measured | 0.0% | 0.0% |
| Invalid moves · Bullet 60s% of moves, lower is better 1 · 1 of 34 measured | 0.0% | 0.0%low reasoning |
| Invalid moves · Lightning 10+1% of moves, lower is better 1 · 1 of 25 measured | 0.0%Ultrafast | 0.0%low reasoning |
| Ladder Elo · Blitz 3+2ladder Elo, higher is better 3 · 6 of 16 measured | 927Ultrafast | 683 |
| Median move time · Blitz 3+2milliseconds, lower is better 6 · 7 of 16 measured | 2.1 sUltrafast | 3.7 s |
| Median move time · Bullet 60smilliseconds, lower is better 19 · 29 of 34 measured | 1.4 sUltrafast (low reasoning) | 2.8 slow reasoning |
| Median move time · Lightning 10+1milliseconds, lower is better 20 · 21 of 25 measured | 1.2 sUltrafast (low reasoning) | 1.3 slow reasoning |
| CatalogBench measured by Spring Prompt · 28 results | ||
| Channel compliance% of products, higher is better 1 · 25 of 26 measured | 100.0% | 98.2% |
| Channel rules broken% of products, lower is better 1 · 23 of 26 measured | 0.0% | 1.8% |
| Channel rules broken · sales brief% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Claims to check% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Claims to check · sales brief% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Conflicts caught% of conflicts, higher is better 1 · 1 of 26 measured | 100.0% | 100.0% |
| Content quality% of checks, higher is better 2 · 9 of 26 measured | 98.6% | 97.7% |
| Content quality · sales brief% of checks, higher is better 2 · 5 of 26 measured | 98.8% | 98.3% |
| Decision accuracy% of decisions, higher is better 2 · 5 of 26 measured | 98.7% | 98.1% |
| Failed outputs% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Failed outputs · sales brief% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Field accuracy% of missing fields, higher is better 6 · 6 of 26 measured | 94.4% | 94.4% |
| Invented values% of filled values, lower is better 19 · 12 of 26 measured | 2.2% | 1.5% |
| Missing UK information% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Missing UK information · sales brief% of products, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Not findable% of products, lower is better 11 · 23 of 26 measured | 4.8% | 10.7% |
| Not findable · sales brief% of products, lower is better 15 · 25 of 26 measured | 5.4% | 13.7% |
| Publish-ready listings% of products, higher is better 2 · 6 of 26 measured | 78.0% | 74.4% |
| Publish-ready listings · sales brief% of products, higher is better 5 · 3 of 26 measured | 72.0% | 74.4% |
| Time per productseconds, lower is better 10 · 9 of 26 measured | 10.2 | 10.1 |
| Unsupported claimsclaims per product, lower is better 1 · 4 of 26 measured | 0.00 | 0.03 |
| Unsupported claims% of products, lower is better 1 · 3 of 26 measured | 0.0% | 0.6% |
| Unsupported claims · sales brief% of products, lower is better 5 · 1 of 26 measured | 3.6% | 1.8% |
| Unsupported claims · sales briefclaims per product, lower is better 2 · 3 of 26 measured | 0.07 | 0.09 |
| Wrong attributes% of products, lower is better 12 · 5 of 26 measured | 16.7% | 14.9% |
| Wrong attributes · sales brief% of products, lower is better 15 · 7 of 26 measured | 19.1% | 14.9% |
| Wrong category or variant% of products, lower is better 5 · 5 of 26 measured | 1.8% | 1.8% |
| Wrong category or variant · sales brief% of products, lower is better 5 · 5 of 26 measured | 1.8% | 1.8% |
| DeckBench measured by Spring Prompt · 14 results | ||
| Caveat dropped% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Clean layout% of tasks, higher is better 3 · 3 of 26 measured | 83.3% | 83.3% |
| Design quality% of the maximum, higher is better 10 · 5 of 26 measured | 71.4% | 80.8% |
| Draft figure quoted% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Findings missing% of tasks, lower is better 1 · 1 of 26 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. 6 · 15 of 26 measured | 82.1% | 48.3% |
| Layout defects% of tasks, lower is better 3 · 3 of 26 measured | 16.7% | 16.7% |
| Misleading metric used% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Presentable decks% of tasks, higher is better 2 · 2 of 26 measured | 50.0% | 50.0% |
| Recommendation late or wrong% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Slides needing work% of tasks, lower is better 13 · 8 of 26 measured | 50.0% | 33.3% |
| Time per deckseconds, lower is better 19 · 16 of 26 measured | 179 | 135 |
| Unsupported claims% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| Unsupported numbers% of tasks, lower is better 1 · 1 of 26 measured | 0.0% | 0.0% |
| SurveyBench measured by Spring Prompt · 17 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. 2 · 6 of 15 measured | 81.4% | 56.2% |
| Leading or double-barrelled questions% of tasks, lower is better 1 · 3 of 15 measured | 0.0% | 11.1% |
| Mistaken next steps% of tasks, lower is better 4 · 4 of 15 measured | 11.1% | 11.1% |
| Numbers right% of answers, higher is better 1 · 1 of 15 measured | 100.0% | 100.0% |
| Objectives unanswerable% of tasks, lower is better 1 · 8 of 15 measured | 0.0% | 11.1% |
| Precision ignored% of tasks, lower is better 1 · 1 of 15 measured | 0.0% | 0.0% |
| Research scorepoints of 100, higher is better 1 · 4 of 15 measured | 97.0 | 92.2 |
| Sound research% of tasks, higher is better 1 · 3 of 15 measured | 66.7% | 33.3% |
| Stakeholder's view presumed% of tasks, lower is better 1 · 6 of 15 measured | 0.0% | 11.1% |
| Survey plan scorepoints of 100, higher is better 1 · 5 of 15 measured | 97.8 | 93.6 |
| Survey structure% of tasks, lower is better 3 · 8 of 15 measured | 11.1% | 22.2% |
| Time per taskseconds, lower is better 12 · 5 of 14 measured | 222 | 92.1 |
| Traps handled% of traps, higher is better 2 · 4 of 15 measured | 95.7% | 94.2% |
| Traps missed% of tasks, lower is better 3 · 3 of 15 measured | 33.3% | 33.3% |
| Unsupported findings% of tasks, lower is better 1 · 1 of 15 measured | 0.0% | 0.0% |
| Wrong numbers% of tasks, lower is better 1 · 1 of 15 measured | 0.0% | 0.0% |
| Wrong recommendation% of tasks, lower is better 3 · 3 of 15 measured | 11.1% | 11.1% |
Compare them with other models
Quick answers
Is GPT-6.1 Sol better than GPT-6.1 Terra?
They score about the same, so pick GPT-6.1 Terra: it costs 2.4 times less per product listing on CatalogBench.
Which is cheaper, GPT-6.1 Sol or GPT-6.1 Terra?
GPT-6.1 Terra is cheaper to run: $0.004 against $0.008 per product listing on CatalogBench (2.4 times less).
Which is better for product listings, GPT-6.1 Sol or GPT-6.1 Terra?
GPT-6.1 Sol ranks 2 of 25 models on sale for turning a sparse product feed and photos into listings that can go live, against 4 for GPT-6.1 Terra.
Which is better for decks from an analysis, GPT-6.1 Sol or GPT-6.1 Terra?
GPT-6.1 Sol ranks 3 of 26 models on sale for turning a finished analysis into a deck you could present as it is, against 7 for GPT-6.1 Terra.
Which is better for user surveys, GPT-6.1 Sol or GPT-6.1 Terra?
GPT-6.1 Sol ranks 3 of 15 models on sale for planning a user survey and reading its results without being misled, against 5 for GPT-6.1 Terra.
Which is better for speed under pressure, GPT-6.1 Sol or GPT-6.1 Terra?
GPT-6.1 Sol ranks 3 of 29 models on sale for good decisions against a real clock (fast chess), against 16 for GPT-6.1 Terra.
Run GPT-6.1 Sol and GPT-6.1 Terra on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.