GPT-5.6 Luna vs GPT-5.6 Terra
GPT-5.6 Terra leads GPT-5.6 Luna on coding, 63.9 to 58.0 on the Vibe Coding Index, and GPT-5.6 Luna costs 10.0x less per output token ($1.20 vs $12.00 per million). Updated October 22, 2026; 117 benchmarks have a result for both.
GPT-5.6 Luna or GPT-5.6 Terra: which to pick
Pick GPT-5.6 Terra for coding: it scores 63.9 on the Vibe Coding Index against 58.0 for GPT-5.6 Luna.
Pick GPT-5.6 Luna for output tokens 10.0x cheaper ($1.20 vs $12.00 per million) and a lower cost per solved task ($0.04 against $0.42).
- Shared results
- 117
- GPT-5.6 Luna only
- 3
- GPT-5.6 Terra only
- 12
- 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 Terra
GPT-5.6 Terra scores 63.9 against 58.0.
6 benchmarks behind each score
Intelligence
AA Intelligence Index
GPT-5.6 Terra
GPT-5.6 Terra leads, 42.1 against 37.3.
Same board, both measured
Agentic work
AA Agentic Index
GPT-5.6 Terra
GPT-5.6 Terra leads, 43.2 against 42.1.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.6 Luna
GPT-5.6 Luna leads, 91.7% against 90.9%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.6 Terra
GPT-5.6 Terra leads, 86.7% against 86.0%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.6 Terra
GPT-5.6 Terra leads, 1268 against 1255.
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 Luna
GPT-5.6 Luna $0.0008, GPT-5.6 Terra $0.0080 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5.6 Luna
GPT-5.6 Luna $0.04, GPT-5.6 Terra $0.42.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.6 Luna 58.0 from 6 benchmarks, GPT-5.6 Terra 63.9 from 6 benchmarks.
- Widest category
- Agentic work, 9.0 points apart on average over 13 shared benchmarks.
- Widest single gap
- IOI (Vals AI): GPT-5.6 Luna 61.8% against GPT-5.6 Terra 87.6%, 25.8 points apart on a 0 to 100 scale.
- Output price
- GPT-5.6 Luna $1.20 against GPT-5.6 Terra $12.00 per million output tokens, 10.0x 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 Luna | Bars | GPT-5.6 Terra | Ties |
|---|---|---|---|---|
| Coding | 5 | 18 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 2 | 11 | ||
| Reasoning | 1 | 16 | ||
| Math | 0 | 6 | ||
| Knowledge | 0 | 11 | 1 tie | |
| Web design | 0 | 8 | 1 tie | |
| General | 4 | 31 | ||
| All shared boards | 12 | 103 | 2 ties |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index518
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| LiveBench coding | 73.6% | 86.2% | GPT-5.6 Terra |
| LMArena WebDev | 1518 | 1522 | GPT-5.6 Terra |
| AA Coding Index | 71.4 | 76.7 | GPT-5.6 Terra |
| LiveCodeBench (Vals AI) | Not measured | 85.9% | Only GPT-5.6 Terra measured |
| SWE-bench Verified (Vals AI) | 93.0% | 95.4% | GPT-5.6 Terra |
| Terminal-Bench 4 (Vals AI) | 11.6% | 22.7% | GPT-5.6 Terra |
| Vibe Code Bench (Vals AI) | 77.1% | 74.6% | GPT-5.6 Luna |
| Code migration (Vals AI) | 44.5% | 47.8% | GPT-5.6 Terra |
| ProgramBench (Vals AI) | 0.0% | 0.5% | GPT-5.6 Terra |
| IOI (Vals AI) | 61.8% | 87.6% | GPT-5.6 Terra |
| Terminal-Bench 2.1 (Vals AI) | 79.0% | 77.5% | GPT-5.6 Luna |
| Vibe Code Bench 1-100 (Vals AI) | 22.6% | 14.8% | GPT-5.6 Luna |
| SkillsBench (Vals AI) | 60.5% | 58.9% | GPT-5.6 Luna |
| CyberBench (Vals AI) | 73.6% | 72.1% | GPT-5.6 Luna |
| SciCode | 53.6% | 55.0% | GPT-5.6 Terra |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 62.9% | Only GPT-5.6 Terra measured |
| FrontierCode | 39.8% | 41.3% | GPT-5.6 Terra |
| WeirdML | 60.9% | 78.3% | GPT-5.6 Terra |
| ALE-Bench | 1667.4 | 1951.4 | GPT-5.6 Terra |
| LMArena Text: coding | 1466 | 1485 | GPT-5.6 Terra |
| LMArena Text: software and IT | 1463 | 1482 | GPT-5.6 Terra |
| LMArena WebDev: React | 1508 | 1513 | GPT-5.6 Terra |
| LMArena WebDev: HTML | 1520 | 1526 | GPT-5.6 Terra |
| LMArena WebDev: from an image | 1517 | 1541 | GPT-5.6 Terra |
| SciCode | 6.2% | 10.8% | GPT-5.6 Terra |
| SWE-Bench Pro V2 | Not measured | 92.4% | Only GPT-5.6 Terra measured |
| Neither model has a result on the other 17 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| AA Intelligence Index | 37.3 | 42.1 | GPT-5.6 Terra |
| Epoch Capabilities Index | 156.5 | 160.1 | GPT-5.6 Terra |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index211
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| AA Agentic Index | 42.1 | 43.2 | GPT-5.6 Terra |
| Design Arena: agents building web apps | Not measured | 1120 | Only GPT-5.6 Terra measured |
| Design Arena: agents building full-stack apps | Not measured | 1061 | Only GPT-5.6 Terra measured |
| Design Arena: agents building mobile apps | Not measured | 1110 | Only GPT-5.6 Terra measured |
| Design Arena: agents building Android apps | Not measured | 1060 | Only GPT-5.6 Terra measured |
| Design Arena: agents building games | Not measured | 1083 | Only GPT-5.6 Terra measured |
| Design Arena: HTML slides | Not measured | 1047 | Only GPT-5.6 Terra measured |
| Design Arena: PowerPoint slides | Not measured | 1104 | Only GPT-5.6 Terra measured |
| LiveBench research | 51.0% | 60.4% | GPT-5.6 Terra |
| LiveBench browser | 68.0% | 84.0% | GPT-5.6 Terra |
| LiveBench long agentic | 30.4% | 48.4% | GPT-5.6 Terra |
| SREBench (Vals AI) | 1.5% | 9.5% | GPT-5.6 Terra |
| Finance Agent (Vals AI) | 55.0% | 54.4% | GPT-5.6 Luna |
| Tax Agent Bench (Vals AI) | 24.8% | 30.0% | GPT-5.6 Terra |
| Legal Research Bench (Vals AI) | 36.5% | 41.4% | GPT-5.6 Terra |
| Excel Modeling Benchmark (Vals AI) | 67.1% | 66.2% | GPT-5.6 Luna |
| BioMysteryBench (Vals AI) | 61.5% | Not measured | Only GPT-5.6 Luna measured |
| Terminal-Bench-Science (Vals AI) | 0.0% | 10.0% | GPT-5.6 Terra |
| Vals Science Index | 25.3% | Not measured | Only GPT-5.6 Luna measured |
| τ²-Bench Telecom | Not measured | 86.3% | Only GPT-5.6 Terra measured |
| APEX-Agents | 43.0% | 58.2% | GPT-5.6 Terra |
| BALROG | 45.6% | 53.2% | GPT-5.6 Terra |
| Vending-Bench 2 | $4,095 | $7,343 | GPT-5.6 Terra |
| Neither model has a result on the other 19 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index116
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | 90.9% | GPT-5.6 Luna |
| LiveBench reasoning | 85.6% | 90.6% | GPT-5.6 Terra |
| MysteryMechanism (Vals AI) | 14.4% | Not measured | Only GPT-5.6 Luna measured |
| MMMU Pro (Vals AI) | 85.0% | 86.5% | GPT-5.6 Terra |
| GPQA Diamond (Artificial Analysis run) | 91.1% | 92.5% | GPT-5.6 Terra |
| Humanity's Last Exam (Artificial Analysis run) | 39.5% | 42.9% | GPT-5.6 Terra |
| GPQA Diamond (Epoch AI run) | 91.6% | 93.3% | GPT-5.6 Terra |
| ARC-AGI | 88.0% | 96.5% | GPT-5.6 Terra |
| ARC-AGI-2 | 59.5% | 83.9% | GPT-5.6 Terra |
| SimpleBench | 46.8% | 48.9% | GPT-5.6 Terra |
| Chess puzzles | 43.0% | 47.0% | GPT-5.6 Terra |
| Mystery game puzzles | 21.0% | 35.0% | GPT-5.6 Terra |
| LMArena Text: hard prompts | 1451 | 1469 | GPT-5.6 Terra |
| LMArena Text: expert | 1478 | 1496 | GPT-5.6 Terra |
| LMArena Text: hard prompts in English | 1460 | 1465 | GPT-5.6 Terra |
| LMArena Vision: diagrams | 1273 | 1295 | GPT-5.6 Terra |
| LMArena Vision: homework | 1282 | 1312 | GPT-5.6 Terra |
| Kagi LLM Benchmark | 49.1% | 51.3% | GPT-5.6 Terra |
| Neither model has a result on the other 5 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.3% | 99.7% | GPT-5.6 Terra |
| FrontierMath, tiers 1 to 3 | 82.1% | 86.0% | GPT-5.6 Terra |
| FrontierMath, tier 4 | 61.0% | 70.7% | GPT-5.6 Terra |
| ProofBench | 60.0% | 74.0% | GPT-5.6 Terra |
| LMArena Text: math | 1459 | 1465 | GPT-5.6 Terra |
| LMArena Text: mathematical work | 1470 | 1478 | GPT-5.6 Terra |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index011
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| MMLU Pro | 86.0% | 86.7% | GPT-5.6 Terra |
| LegalBench (Vals AI) | 84.0% | 85.1% | GPT-5.6 Terra |
| CorpFin (Vals AI) | 64.2% | 65.3% | GPT-5.6 Terra |
| TaxEval (Vals AI) | 76.2% | 76.2% | Tie |
| MortgageTax (Vals AI) | 67.3% | 67.3% | GPT-5.6 Terra |
| MedCode (Vals AI) | 42.4% | 43.4% | GPT-5.6 Terra |
| SimpleQA Verified | 41.0% | 43.2% | GPT-5.6 Terra |
| LMArena Text: science | 1433 | 1452 | GPT-5.6 Terra |
| LMArena Text: medicine and healthcare | 1399 | 1430 | GPT-5.6 Terra |
| LMArena Text: legal and government | 1446 | 1461 | GPT-5.6 Terra |
| LMArena Text: business and finance | 1439 | 1448 | GPT-5.6 Terra |
| FACTS Benchmark Suite | 35.0% | 52.0% | GPT-5.6 Terra |
| Neither model has a result on the other 8 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| Design Arena: websites | 1255 | 1268 | GPT-5.6 Terra |
| Design Arena: UI components | 1248 | 1256 | GPT-5.6 Terra |
| Design Arena: game dev | 1195 | 1229 | GPT-5.6 Terra |
| Design Arena: data viz | 1212 | 1240 | GPT-5.6 Terra |
| Design Arena: 3D | 1121 | 1134 | GPT-5.6 Terra |
| Design Arena: all code categories | 1230 | 1240 | GPT-5.6 Terra |
| Design Arena: SVG | 1193 | 1219 | GPT-5.6 Terra |
| Design Arena: ASCII art | 1142 | 1178 | GPT-5.6 Terra |
| LiveBench design | 76.0% | 76.0% | Tie |
GeneralLike-for-likePublic boards, shown as published, kept out of the index431
| Benchmark | GPT-5.6 Luna | GPT-5.6 Terra | Leader |
|---|---|---|---|
| LiveBench data analysis | 78.0% | 79.3% | GPT-5.6 Terra |
| Public Benefits Bench (Vals AI) | 61.2% | 62.4% | GPT-5.6 Terra |
| SAGE (Vals AI) | 44.2% | 47.0% | GPT-5.6 Terra |
| MedScribe (Vals AI) | 84.4% | 82.9% | GPT-5.6 Luna |
| Vals Multimodal Index | 69.1% | 65.1% | GPT-5.6 Luna |
| IFBench | Not measured | 71.2% | Only GPT-5.6 Terra measured |
| Long context reasoning (AA-LCR) | 83.7% | 83.0% | GPT-5.6 Luna |
| DTBench | 89.1% | 93.3% | GPT-5.6 Terra |
| LMCA | 48.5% | 54.9% | GPT-5.6 Terra |
| Surface Evolver Bench | 61.9% | 83.8% | GPT-5.6 Terra |
| Furniture assembly | 42.5% | 54.2% | GPT-5.6 Terra |
| LMArena Text | 1431 | 1448 | GPT-5.6 Terra |
| LMArena Text: instruction following | 1437 | 1456 | GPT-5.6 Terra |
| LMArena Text: creative writing | 1396 | 1413 | GPT-5.6 Terra |
| LMArena Vision | 1257 | 1271 | GPT-5.6 Terra |
| LMArena Text: English | 1441 | 1449 | GPT-5.6 Terra |
| LMArena Text: non-English | 1417 | 1441 | GPT-5.6 Terra |
| LMArena Text: Chinese | 1470 | 1514 | GPT-5.6 Terra |
| LMArena Text: German | 1455 | 1460 | GPT-5.6 Terra |
| LMArena Text: Japanese | 1415 | 1454 | GPT-5.6 Terra |
| LMArena Text: Korean | 1415 | 1427 | GPT-5.6 Terra |
| LMArena Text: French | 1458 | 1466 | GPT-5.6 Terra |
| LMArena Text: Spanish | 1446 | 1451 | GPT-5.6 Terra |
| LMArena Text: Russian | 1427 | 1452 | GPT-5.6 Terra |
| LMArena Text: Polish | 1405 | 1446 | GPT-5.6 Terra |
| LMArena Text: multi-turn | 1434 | 1453 | GPT-5.6 Terra |
| LMArena Text: longer queries | 1436 | 1454 | GPT-5.6 Terra |
| LMArena Text: writing and language | 1419 | 1431 | GPT-5.6 Terra |
| LMArena Text: entertainment and media | 1402 | 1418 | GPT-5.6 Terra |
| LMArena Vision: English | 1263 | 1272 | GPT-5.6 Terra |
| LMArena Vision: Chinese | 1294 | 1337 | GPT-5.6 Terra |
| LMArena Vision: reading text | 1264 | 1278 | GPT-5.6 Terra |
| LMArena Vision: humor | 1190 | 1236 | GPT-5.6 Terra |
| LMArena Vision: creative writing | 1257 | 1248 | GPT-5.6 Luna |
| LMArena Document | 1457 | 1472 | GPT-5.6 Terra |
| EQ-Bench 4 | 1156 | 1234 | GPT-5.6 Terra |
| Neither model has a result on the other 12 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.6 Luna | 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 Luna costs less
Repository review
50K input + 3K output tokens
GPT-5.6 Luna costs less
Agent loop
200K input + 10K output tokens
GPT-5.6 Luna 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 Luna | 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 | $0.20 | $2.00 |
| Output $/Mtok | $1.20 | $12.00 |
| Coding index | 58.0 | 63.9 |
| Independent measurements behind it | 6 | 6 |
| Cost per solved task | $0.04 | $0.42 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.6 Luna better than GPT-5.6 Terra for coding?
Pick GPT-5.6 Terra for coding: it scores 63.9 on the Vibe Coding Index against 58.0 for GPT-5.6 Luna. Pick GPT-5.6 Luna for output tokens 10.0x cheaper ($1.20 vs $12.00 per million) and a lower cost per solved task ($0.04 against $0.42).
Which is cheaper, GPT-5.6 Luna or GPT-5.6 Terra?
GPT-5.6 Luna costs $0.20 per million input tokens and $1.20 per million output tokens; GPT-5.6 Terra costs $2.00 and $12.00. Per solved SWE-bench Verified task, GPT-5.6 Luna costs $0.04 and GPT-5.6 Terra $0.42.
Which has the bigger context window, GPT-5.6 Luna or GPT-5.6 Terra?
Neither: both take 1.05M tokens.
How many benchmarks is the GPT-5.6 Luna vs GPT-5.6 Terra comparison based on?
117 benchmarks have a result for both models as of October 22, 2026. GPT-5.6 Luna'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.