GPT-5.6 Sol vs GPT-5.6 Terra
GPT-5.6 Sol leads GPT-5.6 Terra on coding, 73.5 to 63.9 on the Vibe Coding Index, and GPT-5.6 Terra costs 1.7x less per output token ($12.00 vs $20.00 per million). Updated October 22, 2026; 126 benchmarks have a result for both.
GPT-5.6 Sol or GPT-5.6 Terra: which to pick
Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 63.9 for GPT-5.6 Terra.
Pick GPT-5.6 Terra for output tokens 1.7x cheaper ($12.00 vs $20.00 per million) and a lower cost per solved task ($0.42 against $1.20).
- Shared results
- 126
- GPT-5.6 Sol only
- 13
- GPT-5.6 Terra only
- 1
- Like-for-like categories
- 8 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
GPT-5.6 Sol
GPT-5.6 Sol scores 73.5 against 63.9.
6 benchmarks behind each score
Intelligence
AA Intelligence Index
GPT-5.6 Sol
GPT-5.6 Sol leads, 47.0 against 42.1.
Same board, both measured
Agentic work
AA Agentic Index
GPT-5.6 Sol
GPT-5.6 Sol leads, 50.2 against 43.2.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.6 Sol
GPT-5.6 Sol leads, 95.2% against 90.9%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.6 Sol
GPT-5.6 Sol leads, 89.1% against 86.7%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.6 Sol
GPT-5.6 Sol leads, 1321 against 1268.
Same board, both measured
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-5.6 Terra
GPT-5.6 Sol $0.014, GPT-5.6 Terra $0.0080 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5.6 Terra
GPT-5.6 Sol $1.20, GPT-5.6 Terra $0.42.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.6 Sol 73.5 from 6 benchmarks, GPT-5.6 Terra 63.9 from 6 benchmarks.
- Widest category
- Web design, 13.5 points apart on average over 9 shared benchmarks.
- Widest single gap
- Design Arena: 3D: GPT-5.6 Sol 1384 against GPT-5.6 Terra 1134, 31.3 points apart on a 0 to 100 scale.
- Output price
- GPT-5.6 Sol $20.00 against GPT-5.6 Terra $12.00 per million output tokens, 1.7x apart.
- Context window
- Both take 1.05M tokens.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-5.6 Sol | Bars | GPT-5.6 Terra | Ties |
|---|---|---|---|---|
| Coding | 22 | 2 | 1 tie | |
| Intelligence | 2 | 0 | ||
| Agentic work | 17 | 3 | ||
| Reasoning | 16 | 1 | ||
| Math | 6 | 0 | ||
| Knowledge | 9 | 3 | ||
| Web design | 9 | 0 | ||
| General | 33 | 2 | ||
| All shared boards | 114 | 11 | 1 tie |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index222
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| LiveBench coding | 89.7% | 86.2% | GPT-5.6 Sol |
| LMArena WebDev | 1616 | 1522 | GPT-5.6 Sol |
| AA Coding Index | 77.4 | 76.7 | GPT-5.6 Sol |
| LiveCodeBench (Vals AI) | 82.6% | 85.9% | GPT-5.6 Terra |
| SWE-bench Verified (Vals AI) | 96.2% | 95.4% | GPT-5.6 Sol |
| Terminal-Bench 4 (Vals AI) | 37.9% | 22.7% | GPT-5.6 Sol |
| Vibe Code Bench (Vals AI) | 80.5% | 74.6% | GPT-5.6 Sol |
| Code migration (Vals AI) | 52.9% | 47.8% | GPT-5.6 Sol |
| ProgramBench (Vals AI) | 1.5% | 0.5% | GPT-5.6 Sol |
| IOI (Vals AI) | 91.2% | 87.6% | GPT-5.6 Sol |
| Terminal-Bench 2.1 (Vals AI) | 85.8% | 77.5% | GPT-5.6 Sol |
| Vibe Code Bench 1-100 (Vals AI) | 20.0% | 14.8% | GPT-5.6 Sol |
| SkillsBench (Vals AI) | 54.1% | 58.9% | GPT-5.6 Terra |
| CyberBench (Vals AI) | 76.3% | 72.1% | GPT-5.6 Sol |
| SciCode | 57.8% | 55.0% | GPT-5.6 Sol |
| Terminal-Bench Hard (Artificial Analysis run) | 65.9% | 62.9% | GPT-5.6 Sol |
| FrontierCode | 47.5% | 41.3% | GPT-5.6 Sol |
| FrontierSWE | 32.2% | Not measured | Only GPT-5.6 Sol measured |
| GSO-Bench | 76.5% | Not measured | Only GPT-5.6 Sol measured |
| WeirdML | 88.8% | 78.3% | GPT-5.6 Sol |
| MirrorCode | 20.0% | Not measured | Only GPT-5.6 Sol measured |
| ALE-Bench | 2176.9 | 1951.4 | GPT-5.6 Sol |
| LMArena Text: coding | 1496 | 1485 | GPT-5.6 Sol |
| LMArena Text: software and IT | 1489 | 1482 | GPT-5.6 Sol |
| LMArena WebDev: React | 1620 | 1513 | GPT-5.6 Sol |
| LMArena WebDev: HTML | 1618 | 1526 | GPT-5.6 Sol |
| LMArena WebDev: from an image | 1609 | 1541 | GPT-5.6 Sol |
| SciCode | 10.8% | 10.8% | Tie |
| Neither model has a result on the other 11 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| AA Intelligence Index | 47.0 | 42.1 | GPT-5.6 Sol |
| Epoch Capabilities Index | 162.8 | 160.1 | GPT-5.6 Sol |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index173
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| AA Agentic Index | 50.2 | 43.2 | GPT-5.6 Sol |
| Design Arena: agents building web apps | 1245 | 1120 | GPT-5.6 Sol |
| Design Arena: agents building full-stack apps | 1261 | 1061 | GPT-5.6 Sol |
| Design Arena: agents building mobile apps | 1248 | 1111 | GPT-5.6 Sol |
| Design Arena: agents building Android apps | 1240 | 1061 | GPT-5.6 Sol |
| Design Arena: agents building Godot games | 1260 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: agents building games | 1266 | 1083 | GPT-5.6 Sol |
| Design Arena: HTML slides | 1236 | 1057 | GPT-5.6 Sol |
| Design Arena: PowerPoint slides | 1217 | 1107 | GPT-5.6 Sol |
| LiveBench research | 54.5% | 60.4% | GPT-5.6 Terra |
| LiveBench browser | 88.0% | 84.0% | GPT-5.6 Sol |
| LiveBench long agentic | 48.5% | 48.4% | GPT-5.6 Sol |
| SREBench (Vals AI) | 30.5% | 9.5% | GPT-5.6 Sol |
| Finance Agent (Vals AI) | 53.8% | 54.4% | GPT-5.6 Terra |
| Tax Agent Bench (Vals AI) | 31.1% | 30.0% | GPT-5.6 Sol |
| Legal Research Bench (Vals AI) | 48.1% | 41.4% | GPT-5.6 Sol |
| Excel Modeling Benchmark (Vals AI) | 72.3% | 66.2% | GPT-5.6 Sol |
| BioMysteryBench (Vals AI) | 71.1% | Not measured | Only GPT-5.6 Sol measured |
| Terminal-Bench-Science (Vals AI) | 20.0% | 10.0% | GPT-5.6 Sol |
| Vals Science Index | 41.5% | Not measured | Only GPT-5.6 Sol measured |
| Time Horizon Index (Vals AI) | 23.8% | Not measured | Only GPT-5.6 Sol measured |
| CUA-bench (Vals AI) | 8.3% | Not measured | Only GPT-5.6 Sol measured |
| τ²-Bench Telecom | 85.1% | 86.3% | GPT-5.6 Terra |
| APEX-Agents | Not measured | 58.2% | Only GPT-5.6 Terra measured |
| OSWorld 2.0 | 27.3% | Not measured | Only GPT-5.6 Sol measured |
| BALROG | 60.0% | 53.2% | GPT-5.6 Sol |
| PostTrainBench | 36.2% | Not measured | Only GPT-5.6 Sol measured |
| Vending-Bench 2 | $9,619 | $7,343 | GPT-5.6 Sol |
| LMArena Search | 1261 | Not measured | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 12 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index161
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| GPQA Diamond | 95.2% | 90.9% | GPT-5.6 Sol |
| LiveBench reasoning | 91.7% | 90.6% | GPT-5.6 Sol |
| MysteryMechanism (Vals AI) | 33.3% | Not measured | Only GPT-5.6 Sol measured |
| MMMU Pro (Vals AI) | 88.8% | 86.5% | GPT-5.6 Sol |
| GPQA Diamond (Artificial Analysis run) | 94.1% | 92.5% | GPT-5.6 Sol |
| Humanity's Last Exam (Artificial Analysis run) | 49.5% | 42.9% | GPT-5.6 Sol |
| GPQA Diamond (Epoch AI run) | 93.5% | 93.3% | GPT-5.6 Sol |
| ARC-AGI | 97.5% | 96.5% | GPT-5.6 Sol |
| ARC-AGI-2 | 92.5% | 83.9% | GPT-5.6 Sol |
| SimpleBench | 64.8% | 48.9% | GPT-5.6 Sol |
| Chess puzzles | 59.0% | 47.0% | GPT-5.6 Sol |
| Mystery game puzzles | 58.0% | 35.0% | GPT-5.6 Sol |
| EBR-bench | 44.8% | Not measured | Only GPT-5.6 Sol measured |
| LMArena Text: hard prompts | 1483 | 1469 | GPT-5.6 Sol |
| LMArena Text: expert | 1517 | 1496 | GPT-5.6 Sol |
| LMArena Text: hard prompts in English | 1477 | 1465 | GPT-5.6 Sol |
| LMArena Vision: diagrams | 1305 | 1295 | GPT-5.6 Sol |
| LMArena Vision: homework | 1305 | 1312 | GPT-5.6 Terra |
| Kagi LLM Benchmark | 67.0% | 51.3% | GPT-5.6 Sol |
| Neither model has a result on the other 3 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 99.7% | GPT-5.6 Sol |
| FrontierMath, tiers 1 to 3 | 89.1% | 86.0% | GPT-5.6 Sol |
| FrontierMath, tier 4 | 82.9% | 70.7% | GPT-5.6 Sol |
| ProofBench | 83.0% | 74.0% | GPT-5.6 Sol |
| LMArena Text: math | 1473 | 1465 | GPT-5.6 Sol |
| LMArena Text: mathematical work | 1496 | 1478 | GPT-5.6 Sol |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index93
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| MMLU Pro | 89.1% | 86.7% | GPT-5.6 Sol |
| LegalBench (Vals AI) | 87.0% | 85.1% | GPT-5.6 Sol |
| CorpFin (Vals AI) | 64.4% | 65.3% | GPT-5.6 Terra |
| TaxEval (Vals AI) | 74.8% | 76.2% | GPT-5.6 Terra |
| MortgageTax (Vals AI) | 67.3% | 67.3% | GPT-5.6 Terra |
| MedCode (Vals AI) | 44.0% | 43.4% | GPT-5.6 Sol |
| SimpleQA Verified | 69.7% | 43.2% | GPT-5.6 Sol |
| LMArena Text: science | 1455 | 1452 | GPT-5.6 Sol |
| LMArena Text: medicine and healthcare | 1431 | 1430 | GPT-5.6 Sol |
| LMArena Text: legal and government | 1476 | 1461 | GPT-5.6 Sol |
| LMArena Text: business and finance | 1457 | 1448 | GPT-5.6 Sol |
| FACTS Benchmark Suite | 52.4% | 52.0% | GPT-5.6 Sol |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index90
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| Design Arena: websites | 1321 | 1268 | GPT-5.6 Sol |
| Design Arena: UI components | 1335 | 1256 | GPT-5.6 Sol |
| Design Arena: game dev | 1367 | 1229 | GPT-5.6 Sol |
| Design Arena: data viz | 1323 | 1240 | GPT-5.6 Sol |
| Design Arena: 3D | 1384 | 1134 | GPT-5.6 Sol |
| Design Arena: all code categories | 1341 | 1240 | GPT-5.6 Sol |
| Design Arena: SVG | 1315 | 1219 | GPT-5.6 Sol |
| Design Arena: ASCII art | 1304 | 1178 | GPT-5.6 Sol |
| LiveBench design | 81.6% | 76.0% | GPT-5.6 Sol |
GeneralLike-for-likePublic boards, shown as published, kept out of the index332
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| LiveBench data analysis | 79.8% | 79.3% | GPT-5.6 Sol |
| Public Benefits Bench (Vals AI) | 66.5% | 62.4% | GPT-5.6 Sol |
| SAGE (Vals AI) | 52.6% | 47.0% | GPT-5.6 Sol |
| MedScribe (Vals AI) | 85.2% | 82.9% | GPT-5.6 Sol |
| Vals Multimodal Index | 72.6% | 65.1% | GPT-5.6 Sol |
| IFBench | 72.7% | 71.2% | GPT-5.6 Sol |
| Long context reasoning (AA-LCR) | 84.0% | 83.0% | GPT-5.6 Sol |
| DTBench | 95.7% | 93.3% | GPT-5.6 Sol |
| LMCA | 58.5% | 54.9% | GPT-5.6 Sol |
| Surface Evolver Bench | 93.1% | 83.8% | GPT-5.6 Sol |
| Furniture assembly | 56.7% | 54.2% | GPT-5.6 Sol |
| LMArena Text | 1457 | 1448 | GPT-5.6 Sol |
| LMArena Text: instruction following | 1483 | 1456 | GPT-5.6 Sol |
| LMArena Text: creative writing | 1449 | 1413 | GPT-5.6 Sol |
| LMArena Vision | 1281 | 1271 | GPT-5.6 Sol |
| LMArena Text: English | 1457 | 1449 | GPT-5.6 Sol |
| LMArena Text: non-English | 1451 | 1441 | GPT-5.6 Sol |
| LMArena Text: Chinese | 1525 | 1514 | GPT-5.6 Sol |
| LMArena Text: German | 1476 | 1460 | GPT-5.6 Sol |
| LMArena Text: Japanese | 1472 | 1454 | GPT-5.6 Sol |
| LMArena Text: Korean | 1442 | 1427 | GPT-5.6 Sol |
| LMArena Text: French | 1472 | 1466 | GPT-5.6 Sol |
| LMArena Text: Spanish | 1443 | 1451 | GPT-5.6 Terra |
| LMArena Text: Russian | 1471 | 1452 | GPT-5.6 Sol |
| LMArena Text: Polish | 1459 | 1446 | GPT-5.6 Sol |
| LMArena Text: multi-turn | 1461 | 1453 | GPT-5.6 Sol |
| LMArena Text: longer queries | 1481 | 1454 | GPT-5.6 Sol |
| LMArena Text: writing and language | 1462 | 1431 | GPT-5.6 Sol |
| LMArena Text: entertainment and media | 1450 | 1418 | GPT-5.6 Sol |
| LMArena Vision: English | 1280 | 1272 | GPT-5.6 Sol |
| LMArena Vision: Chinese | 1313 | 1337 | GPT-5.6 Terra |
| LMArena Vision: reading text | 1290 | 1278 | GPT-5.6 Sol |
| LMArena Vision: humor | 1246 | 1236 | GPT-5.6 Sol |
| LMArena Vision: creative writing | 1283 | 1248 | GPT-5.6 Sol |
| LMArena Document | 1483 | 1472 | GPT-5.6 Sol |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.6 Sol | GPT-5.6 Terra | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-5.6 Terra costs less
Repository review
50K input + 3K output tokens
GPT-5.6 Terra costs less
Agent loop
200K input + 10K output tokens
GPT-5.6 Terra costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-5.6 Sol | OpenAI GPT-5.6 Terra |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | July 9, 2026 | July 9, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $4.00 | $2.00 |
| Output $/Mtok | $20.00 | $12.00 |
| Coding index | 73.5 | 63.9 |
| Independent measurements behind it | 6 | 6 |
| Cost per solved task | $1.20 | $0.42 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.6 Sol better than GPT-5.6 Terra for coding?
Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 63.9 for GPT-5.6 Terra. Pick GPT-5.6 Terra for output tokens 1.7x cheaper ($12.00 vs $20.00 per million) and a lower cost per solved task ($0.42 against $1.20).
Which is cheaper, GPT-5.6 Sol or GPT-5.6 Terra?
GPT-5.6 Sol costs $4.00 per million input tokens and $20.00 per million output tokens; GPT-5.6 Terra costs $2.00 and $12.00. Per solved SWE-bench Verified task, GPT-5.6 Sol costs $1.20 and GPT-5.6 Terra $0.42.
Which has the bigger context window, GPT-5.6 Sol or GPT-5.6 Terra?
Neither: both take 1.05M tokens.
How many benchmarks is the GPT-5.6 Sol vs GPT-5.6 Terra comparison based on?
126 benchmarks have a result for both models as of October 22, 2026. GPT-5.6 Sol's coding index rests on 6 independent benchmarks and GPT-5.6 Terra's on 6. Where only one model has a result, the page says so instead of naming a winner.