GPT-5.6 Sol vs GPT-6 Luna
GPT-5.6 Sol leads GPT-6 Luna on coding, 73.5 to 57.3 on the Vibe Coding Index, and GPT-6 Luna costs 40.0x less per output token ($0.50 vs $20.00 per million). Updated October 22, 2026; 98 benchmarks have a result for both.
GPT-5.6 Sol or GPT-6 Luna: which to pick
Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 57.3 for GPT-6 Luna.
Pick GPT-6 Luna for output tokens 40.0x cheaper ($0.50 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or MedCode (Vals AI), where it leads.
- Shared results
- 98
- GPT-5.6 Sol only
- 41
- GPT-6 Luna only
- 5
- 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 57.3.
6 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.6 Sol
GPT-5.6 Sol leads, 47.0 against 38.1.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only GPT-5.6 Sol has a AA Agentic Index result, 50.2.
Directional only
Reasoning
GPQA Diamond
No call
Only GPT-5.6 Sol has a GPQA Diamond result, 95.2%.
Directional only
Knowledge
MMLU Pro
No call
Only GPT-5.6 Sol has a MMLU Pro result, 89.1%.
Directional only
Web design
Design Arena: websites
GPT-5.6 Sol
GPT-5.6 Sol leads, 1321 against 1271.
Same board, both measured
General
Vals Index
No call
Only GPT-6 Luna has a Vals Index result, 52.3%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6 Luna
GPT-5.6 Sol $0.014, GPT-6 Luna $0.0003 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only GPT-5.6 Sol has the number, $1.20.
Directional only
At a glance
- Coding index
- GPT-5.6 Sol 73.5 from 6 benchmarks, GPT-6 Luna 57.3 from 5 benchmarks.
- Widest category
- Reasoning, 16.1 points apart on average over 13 shared benchmarks.
- Widest single gap
- Mystery game puzzles: GPT-5.6 Sol 58.0% against GPT-6 Luna 7.0%, 51.0 points apart on a 0 to 100 scale.
- Output price
- GPT-5.6 Sol $20.00 against GPT-6 Luna $0.50 per million output tokens, 40.0x 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 Sol | Bars | GPT-6 Luna | Ties |
|---|---|---|---|---|
| Coding | 16 | 1 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 14 | 0 | ||
| Reasoning | 13 | 0 | ||
| Math | 6 | 0 | ||
| Knowledge | 5 | 2 | ||
| Web design | 8 | 1 | ||
| General | 30 | 0 | ||
| All shared boards | 94 | 4 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index161
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| LiveBench coding | 89.7% | 84.5% | GPT-5.6 Sol |
| LMArena WebDev | 1616 | 1578 | GPT-5.6 Sol |
| AA Coding Index | 77.4 | Not measured | Only GPT-5.6 Sol measured |
| LiveCodeBench (Vals AI) | 82.6% | Not measured | Only GPT-5.6 Sol measured |
| SWE-bench Verified (Vals AI) | 96.2% | Not measured | Only GPT-5.6 Sol measured |
| Terminal-Bench 4 (Vals AI) | 37.9% | 13.6% | GPT-5.6 Sol |
| Vibe Code Bench (Vals AI) | 80.5% | 81.7% | GPT-6 Luna |
| Code migration (Vals AI) | 52.9% | 42.5% | GPT-5.6 Sol |
| ProgramBench (Vals AI) | 1.5% | 0.5% | GPT-5.6 Sol |
| IOI (Vals AI) | 91.2% | 55.6% | GPT-5.6 Sol |
| Terminal-Bench 2.1 (Vals AI) | 85.8% | 73.0% | GPT-5.6 Sol |
| SWE-bench Multimodal (Vals AI) | Not measured | 37.5% | Only GPT-6 Luna measured |
| Vibe Code Bench 1-100 (Vals AI) | 20.0% | Not measured | Only GPT-5.6 Sol measured |
| SkillsBench (Vals AI) | 54.1% | Not measured | Only GPT-5.6 Sol measured |
| CyberBench (Vals AI) | 76.3% | 76.3% | GPT-5.6 Sol |
| SciCode | 57.8% | 54.6% | GPT-5.6 Sol |
| Terminal-Bench Hard (Artificial Analysis run) | 65.9% | Not measured | Only GPT-5.6 Sol measured |
| FrontierCode | 47.5% | 42.4% | 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% | Not measured | Only GPT-5.6 Sol measured |
| MirrorCode | 20.0% | Not measured | Only GPT-5.6 Sol measured |
| ALE-Bench | 2176.9 | 1576.9 | GPT-5.6 Sol |
| LMArena Text: coding | 1496 | 1437 | GPT-5.6 Sol |
| LMArena Text: software and IT | 1489 | 1423 | GPT-5.6 Sol |
| LMArena WebDev: React | 1620 | 1573 | GPT-5.6 Sol |
| LMArena WebDev: HTML | 1618 | 1595 | GPT-5.6 Sol |
| LMArena WebDev: from an image | 1609 | 1571 | GPT-5.6 Sol |
| SciCode | 10.8% | Not measured | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 10 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| AA Intelligence Index | 47.0 | 38.1 | GPT-5.6 Sol |
| Epoch Capabilities Index | 162.8 | 156.8 | GPT-5.6 Sol |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index140
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| AA Agentic Index | 50.2 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: agents building web apps | 1245 | 1191 | GPT-5.6 Sol |
| Design Arena: agents building full-stack apps | 1261 | 1158 | GPT-5.6 Sol |
| Design Arena: agents building mobile apps | 1248 | 1134 | GPT-5.6 Sol |
| Design Arena: agents building Android apps | 1240 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: agents building Godot games | 1260 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: agents building games | 1266 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: HTML slides | 1236 | Not measured | Only GPT-5.6 Sol measured |
| Design Arena: PowerPoint slides | 1215 | Not measured | Only GPT-5.6 Sol measured |
| LiveBench research | 54.5% | 43.5% | GPT-5.6 Sol |
| LiveBench browser | 88.0% | 72.0% | GPT-5.6 Sol |
| LiveBench long agentic | 48.5% | 48.1% | GPT-5.6 Sol |
| SREBench (Vals AI) | 30.5% | 2.7% | GPT-5.6 Sol |
| Finance Agent (Vals AI) | 53.8% | 49.9% | GPT-5.6 Sol |
| Tax Agent Bench (Vals AI) | 31.1% | 19.9% | GPT-5.6 Sol |
| Legal Research Bench (Vals AI) | 48.1% | 30.3% | GPT-5.6 Sol |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 4.2% | Only GPT-6 Luna measured |
| Excel Modeling Benchmark (Vals AI) | 72.3% | 68.5% | GPT-5.6 Sol |
| BioMysteryBench (Vals AI) | 71.1% | 61.5% | GPT-5.6 Sol |
| Terminal-Bench-Science (Vals AI) | 20.0% | 4.3% | GPT-5.6 Sol |
| Vals Science Index | 41.5% | 28.4% | GPT-5.6 Sol |
| 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% | Not measured | Only GPT-5.6 Sol measured |
| APEX-Agents | Not measured | 44.3% | Only GPT-6 Luna measured |
| OSWorld 2.0 | 27.3% | Not measured | Only GPT-5.6 Sol measured |
| BALROG | 60.0% | Not measured | Only GPT-5.6 Sol measured |
| PostTrainBench | 36.2% | Not measured | Only GPT-5.6 Sol measured |
| SpireClimb | Not measured | 0.0% | Only GPT-6 Luna measured |
| Vending-Bench 2 | $9,619 | Not measured | Only GPT-5.6 Sol measured |
| LMArena Search | 1261 | Not measured | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 10 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index130
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| GPQA Diamond | 95.2% | Not measured | Only GPT-5.6 Sol measured |
| LiveBench reasoning | 91.7% | 81.8% | GPT-5.6 Sol |
| MysteryMechanism (Vals AI) | 33.3% | 19.4% | GPT-5.6 Sol |
| MMMU Pro (Vals AI) | 88.8% | Not measured | Only GPT-5.6 Sol measured |
| GPQA Diamond (Artificial Analysis run) | 94.1% | Not measured | Only GPT-5.6 Sol measured |
| Humanity's Last Exam (Artificial Analysis run) | 49.5% | 38.5% | GPT-5.6 Sol |
| GPQA Diamond (Epoch AI run) | 93.5% | 90.5% | GPT-5.6 Sol |
| ARC-AGI | 97.5% | 86.7% | GPT-5.6 Sol |
| ARC-AGI-2 | 92.5% | 59.3% | GPT-5.6 Sol |
| SimpleBench | 64.8% | Not measured | Only GPT-5.6 Sol measured |
| Chess puzzles | 59.0% | 36.0% | GPT-5.6 Sol |
| Mystery game puzzles | 58.0% | 7.0% | GPT-5.6 Sol |
| EBR-bench | 44.8% | Not measured | Only GPT-5.6 Sol measured |
| LMArena Text: hard prompts | 1483 | 1410 | GPT-5.6 Sol |
| LMArena Text: expert | 1517 | 1442 | GPT-5.6 Sol |
| LMArena Text: hard prompts in English | 1477 | 1410 | GPT-5.6 Sol |
| LMArena Vision: diagrams | 1305 | 1245 | GPT-5.6 Sol |
| LMArena Vision: homework | 1305 | 1261 | GPT-5.6 Sol |
| Kagi LLM Benchmark | 67.0% | Not measured | Only GPT-5.6 Sol measured |
| 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-6 Luna | Leader |
|---|---|---|---|
| OTIS Mock AIME | 100.0% | 98.9% | GPT-5.6 Sol |
| FrontierMath, tiers 1 to 3 | 89.1% | 78.9% | GPT-5.6 Sol |
| FrontierMath, tier 4 | 82.9% | 56.1% | GPT-5.6 Sol |
| ProofBench | 83.0% | 64.0% | GPT-5.6 Sol |
| LMArena Text: math | 1473 | 1409 | GPT-5.6 Sol |
| LMArena Text: mathematical work | 1496 | 1427 | 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 index52
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| MMLU Pro | 89.1% | Not measured | Only GPT-5.6 Sol measured |
| LegalBench (Vals AI) | 87.0% | Not measured | Only GPT-5.6 Sol measured |
| CorpFin (Vals AI) | 64.4% | Not measured | Only GPT-5.6 Sol measured |
| TaxEval (Vals AI) | 74.8% | Not measured | Only GPT-5.6 Sol measured |
| MortgageTax (Vals AI) | 67.3% | Not measured | Only GPT-5.6 Sol measured |
| MedCode (Vals AI) | 44.0% | 44.7% | GPT-6 Luna |
| SimpleQA Verified | 69.7% | 41.4% | GPT-5.6 Sol |
| LMArena Text: science | 1455 | 1386 | GPT-5.6 Sol |
| LMArena Text: medicine and healthcare | 1431 | 1352 | GPT-5.6 Sol |
| LMArena Text: legal and government | 1476 | 1400 | GPT-5.6 Sol |
| LMArena Text: business and finance | 1457 | 1388 | GPT-5.6 Sol |
| FACTS Benchmark Suite | 52.4% | 63.3% | GPT-6 Luna |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index81
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| Design Arena: websites | 1321 | 1271 | GPT-5.6 Sol |
| Design Arena: UI components | 1335 | 1295 | GPT-5.6 Sol |
| Design Arena: game dev | 1367 | 1257 | GPT-5.6 Sol |
| Design Arena: data viz | 1322 | 1216 | GPT-5.6 Sol |
| Design Arena: 3D | 1384 | 1218 | GPT-5.6 Sol |
| Design Arena: all code categories | 1341 | 1270 | GPT-5.6 Sol |
| Design Arena: SVG | 1314 | 1252 | GPT-5.6 Sol |
| Design Arena: ASCII art | 1304 | 1153 | GPT-5.6 Sol |
| LiveBench design | 81.6% | 82.5% | GPT-6 Luna |
GeneralLike-for-likePublic boards, shown as published, kept out of the index300
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | Leader |
|---|---|---|---|
| Vals Index | Not measured | 52.3% | Only GPT-6 Luna measured |
| LiveBench data analysis | 79.8% | 73.4% | GPT-5.6 Sol |
| Public Benefits Bench (Vals AI) | 66.5% | 57.6% | GPT-5.6 Sol |
| SAGE (Vals AI) | 52.6% | 48.1% | GPT-5.6 Sol |
| MedScribe (Vals AI) | 85.2% | 83.7% | GPT-5.6 Sol |
| Vals Multimodal Index | 72.6% | Not measured | Only GPT-5.6 Sol measured |
| IFBench | 72.7% | Not measured | Only GPT-5.6 Sol measured |
| Long context reasoning (AA-LCR) | 84.0% | 83.3% | GPT-5.6 Sol |
| DTBench | 95.7% | 90.1% | GPT-5.6 Sol |
| LMCA | 58.5% | 44.5% | GPT-5.6 Sol |
| Surface Evolver Bench | 93.1% | Not measured | Only GPT-5.6 Sol measured |
| Furniture assembly | 56.7% | 44.2% | GPT-5.6 Sol |
| LMArena Text | 1457 | 1389 | GPT-5.6 Sol |
| LMArena Text: instruction following | 1483 | 1408 | GPT-5.6 Sol |
| LMArena Text: creative writing | 1449 | 1363 | GPT-5.6 Sol |
| LMArena Vision | 1281 | 1216 | GPT-5.6 Sol |
| LMArena Text: English | 1457 | 1393 | GPT-5.6 Sol |
| LMArena Text: non-English | 1451 | 1381 | GPT-5.6 Sol |
| LMArena Text: Chinese | 1525 | 1427 | GPT-5.6 Sol |
| LMArena Text: German | 1476 | 1376 | GPT-5.6 Sol |
| LMArena Text: Japanese | 1472 | 1369 | GPT-5.6 Sol |
| LMArena Text: Korean | 1442 | 1359 | GPT-5.6 Sol |
| LMArena Text: French | 1472 | 1408 | GPT-5.6 Sol |
| LMArena Text: Spanish | 1443 | 1391 | GPT-5.6 Sol |
| LMArena Text: Russian | 1471 | 1392 | GPT-5.6 Sol |
| LMArena Text: Polish | 1459 | 1383 | GPT-5.6 Sol |
| LMArena Text: multi-turn | 1461 | 1392 | GPT-5.6 Sol |
| LMArena Text: longer queries | 1481 | 1408 | GPT-5.6 Sol |
| LMArena Text: writing and language | 1462 | 1384 | GPT-5.6 Sol |
| LMArena Text: entertainment and media | 1450 | 1364 | GPT-5.6 Sol |
| LMArena Vision: English | 1280 | 1204 | GPT-5.6 Sol |
| LMArena Vision: Chinese | 1313 | 1266 | GPT-5.6 Sol |
| LMArena Vision: reading text | 1290 | 1225 | GPT-5.6 Sol |
| LMArena Vision: humor | 1246 | Not measured | Only GPT-5.6 Sol measured |
| LMArena Vision: creative writing | 1283 | 1205 | GPT-5.6 Sol |
| LMArena Document | 1483 | Not measured | Only GPT-5.6 Sol measured |
| Neither model has a result on the other 10 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.6 Sol | GPT-6 Luna | 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-6 Luna costs less
Repository review
50K input + 3K output tokens
GPT-6 Luna costs less
Agent loop
200K input + 10K output tokens
GPT-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 Sol | OpenAI GPT-6 Luna |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | July 9, 2026 | September 22, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $4.00 | $0.10 |
| Output $/Mtok | $20.00 | $0.50 |
| Coding index | 73.5 | 57.3 |
| Independent measurements behind it | 6 | 5 |
| Cost per solved task | $1.20 | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.6 Sol better than GPT-6 Luna for coding?
Pick GPT-5.6 Sol for coding: it scores 73.5 on the Vibe Coding Index against 57.3 for GPT-6 Luna. Pick GPT-6 Luna for output tokens 40.0x cheaper ($0.50 vs $20.00 per million) and work measured by Vibe Code Bench (Vals AI) or MedCode (Vals AI), where it leads.
Which is cheaper, GPT-5.6 Sol or GPT-6 Luna?
GPT-5.6 Sol costs $4.00 per million input tokens and $20.00 per million output tokens; GPT-6 Luna costs $0.10 and $0.50. Cost per solved task is known for GPT-5.6 Sol only.
Which has the bigger context window, GPT-5.6 Sol or GPT-6 Luna?
Neither: both take 1.05M tokens.
How many benchmarks is the GPT-5.6 Sol vs GPT-6 Luna comparison based on?
98 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-6 Luna's on 5. Where only one model has a result, the page says so instead of naming a winner.