GPT-6 Luna vs Inkling 2 Small
GPT-6 Luna leads Inkling 2 Small on coding, 54.9 to 48.9 on the Vibe Coding Index, and GPT-6 Luna costs 2.0x less per output token ($0.50 vs $1.00 per million). Updated November 14, 2026; 48 benchmarks have a result for both.
vs boards won of 48 shared, 2 tied
GPT-6 Luna or Inkling 2 Small: which to pick
Pick GPT-6 Luna for coding: it scores 54.9 on the Vibe Coding Index against 48.9 for Inkling 2 Small.
Pick Inkling 2 Small for work measured by LMArena WebDev or SWE-bench Multimodal (Vals AI), where it leads.
Model AGPT-6 LunaOpenAI54.9/1005 independent benchmarksNo cost per solved taskModel BInkling 2 SmallThinking Machines48.9/1003 independent benchmarksNo cost per solved task
- Shared results
- 48
- GPT-6 Luna only
- 70
- Inkling 2 Small only
- 3
- Like-for-like categories
- 6 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-6 Luna
GPT-6 Luna scores 54.9 against 48.9.
5 and 3 benchmarks behind the scores
Intelligence
AA Intelligence Index
Inkling 2 Small
Inkling 2 Small leads, 30.6 against 27.9.
Same board, both measured
Web design
Design Arena: websites
No call
Only GPT-6 Luna has a Design Arena: websites result, 1268.
Directional only
General
Vals Index
Inkling 2 Small
Inkling 2 Small leads, 56.1% against 52.3%.
Same board, both measured
Long documents
Listed context window
GPT-6 Luna
GPT-6 Luna takes 1.05M tokens against 1.05M.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-6 Luna
GPT-6 Luna $0.0003, Inkling 2 Small $0.0008 per turn.
List prices
At a glance
- Coding index
- GPT-6 Luna 54.9 from 5 benchmarks, Inkling 2 Small 48.9 from 3 benchmarks.
- Widest category
- Agentic work, 8.8 points apart on average over 8 shared benchmarks.
- Widest single gap
- ARC-AGI-2: GPT-6 Luna 59.3% against Inkling 2 Small 81.7%, 22.4 points apart on a 0 to 100 scale.
- Output price
- GPT-6 Luna $0.50 against Inkling 2 Small $1.00 per million output tokens, 2.0x apart.
- Context window
- GPT-6 Luna takes 1.05M tokens, Inkling 2 Small 1.05M.
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-6 Luna | Bars | Inkling 2 Small | Ties |
|---|---|---|---|---|
| Coding | 4 | 8 | 1 tie | |
| Intelligence | 0 | 1 | ||
| Agentic work | 2 | 6 | ||
| Reasoning | 1 | 8 | ||
| Knowledge | 0 | 3 | ||
| General | 3 | 10 | 1 tie | |
| All shared boards | 10 | 36 | 2 ties |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index48
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| LiveBench coding | 70.8% | Not measured | Only GPT-6 Luna measured |
| LMArena WebDev | 1577 | 1647 | Inkling 2 Small |
| Terminal-Bench 4 (Vals AI) | 13.6% | 12.6% | GPT-6 Luna |
| Vibe Code Bench (Vals AI) | 81.7% | 81.5% | GPT-6 Luna |
| Code migration (Vals AI) | 42.5% | Not measured | Only GPT-6 Luna measured |
| ProgramBench (Vals AI) | 0.5% | 0.5% | Tie |
| IOI (Vals AI) | 55.6% | Not measured | Only GPT-6 Luna measured |
| Terminal-Bench 2.1 (Vals AI) | 73.0% | Not measured | Only GPT-6 Luna measured |
| SWE-bench Multimodal (Vals AI) | 37.5% | 43.1% | Inkling 2 Small |
| CyberBench (Vals AI) | 76.3% | 72.7% | GPT-6 Luna |
| SciCode | 54.6% | 52.3% | GPT-6 Luna |
| Terminal-Bench 4.0 (Artificial Analysis run) | 12.6% | 13.1% | Inkling 2 Small |
| FrontierCode 2.0 | 25.2% | 25.4% | Inkling 2 Small |
| FrontierCode | 42.4% | Not measured | Only GPT-6 Luna measured |
| MirrorCode | 26.7% | Not measured | Only GPT-6 Luna measured |
| ALE-Bench | 1576.9 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: coding | 1431 | 1455 | Inkling 2 Small |
| LMArena Text: software and IT | 1422 | 1443 | Inkling 2 Small |
| LMArena WebDev: React | 1574 | 1660 | Inkling 2 Small |
| LMArena WebDev: HTML | 1595 | 1661 | Inkling 2 Small |
| LMArena WebDev: from an image | 1566 | Not measured | Only GPT-6 Luna measured |
| SWE-Bench Pro V2 | Not measured | 97.7% | Only Inkling 2 Small measured |
| GameBench 2 | 55.9 | Not measured | Only GPT-6 Luna measured |
| Terminal-Bench 4.0 | 16.4% | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| AA Intelligence Index | 27.9 | 30.6 | Inkling 2 Small |
| Epoch Capabilities Index | 156.4 | Not measured | Only GPT-6 Luna measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index26
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1150 | Not measured | Only GPT-6 Luna measured |
| Design Arena: agents building full-stack apps | 1138 | Not measured | Only GPT-6 Luna measured |
| Design Arena: agents building mobile apps | 1117 | Not measured | Only GPT-6 Luna measured |
| Design Arena: agents building Android apps | 1176 | Not measured | Only GPT-6 Luna measured |
| Design Arena: spreadsheets | 875 | Not measured | Only GPT-6 Luna measured |
| LiveBench research | 48.5% | Not measured | Only GPT-6 Luna measured |
| LiveBench browser | 70.3% | Not measured | Only GPT-6 Luna measured |
| LiveBench long agentic | 24.7% | Not measured | Only GPT-6 Luna measured |
| SREBench (Vals AI) | 2.7% | 1.9% | GPT-6 Luna |
| Finance Agent (Vals AI) | 49.9% | Not measured | Only GPT-6 Luna measured |
| Tax Agent Bench (Vals AI) | 19.9% | 38.9% | Inkling 2 Small |
| Legal Research Bench (Vals AI) | 30.3% | 46.6% | Inkling 2 Small |
| Harvey's Legal Agent Benchmark (Vals AI) | 4.2% | 21.7% | Inkling 2 Small |
| Excel Modeling Benchmark (Vals AI) | 68.5% | Not measured | Only GPT-6 Luna measured |
| BioMysteryBench (Vals AI) | 61.5% | 70.0% | Inkling 2 Small |
| Terminal-Bench-Science (Vals AI) | 4.3% | 2.9% | GPT-6 Luna |
| Vals Science Index | 28.4% | 32.7% | Inkling 2 Small |
| APEX-Agents | 44.3% | 46.8% | Inkling 2 Small |
| SpireClimb | 0.0% | Not measured | Only GPT-6 Luna measured |
| LMArena Search | 1160 | Not measured | Only GPT-6 Luna measured |
| MCP Atlas | Not measured | 85.4% | Only Inkling 2 Small measured |
| Neither model has a result on the other 23 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index18
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| LiveBench reasoning | 81.8% | Not measured | Only GPT-6 Luna measured |
| MysteryMechanism (Vals AI) | 19.4% | 25.2% | Inkling 2 Small |
| Humanity's Last Exam (Artificial Analysis run) | 38.5% | 44.1% | Inkling 2 Small |
| HLE-Diamond (Artificial Analysis run) | 22.2% | 28.7% | Inkling 2 Small |
| CritPt v2, hard subset (Artificial Analysis run) | 42.5% | 35.0% | GPT-6 Luna |
| GPQA Diamond (Epoch AI run) | 90.5% | Not measured | Only GPT-6 Luna measured |
| ARC-AGI | 86.7% | 96.5% | Inkling 2 Small |
| ARC-AGI-2 | 59.3% | 81.7% | Inkling 2 Small |
| Chess puzzles | 36.0% | Not measured | Only GPT-6 Luna measured |
| Mystery game puzzles | 7.0% | Not measured | Only GPT-6 Luna measured |
| LMArena Text: hard prompts | 1407 | 1421 | Inkling 2 Small |
| LMArena Text: expert | 1437 | 1478 | Inkling 2 Small |
| LMArena Text: hard prompts in English | 1407 | 1424 | Inkling 2 Small |
| LMArena Vision: diagrams | 1235 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: homework | 1252 | Not measured | Only GPT-6 Luna measured |
| GoBench | 1120 | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 10 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| OTIS Mock AIME | 98.9% | Not measured | Only GPT-6 Luna measured |
| FrontierMath, tiers 1 to 3 | 78.9% | Not measured | Only GPT-6 Luna measured |
| FrontierMath, tier 4 | 56.1% | Not measured | Only GPT-6 Luna measured |
| ProofBench | 64.0% | Not measured | Only GPT-6 Luna measured |
| LMArena Text: math | 1411 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: mathematical work | 1425 | Not measured | Only GPT-6 Luna measured |
| MathArena | Not measured | 42.4% | Only Inkling 2 Small measured |
| Neither model has a result on the other 7 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index03
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| MedCode (Vals AI) | 44.7% | Not measured | Only GPT-6 Luna measured |
| AA-Omniscience | -8.3 | 16.1 | Inkling 2 Small |
| SimpleQA Verified | 41.4% | Not measured | Only GPT-6 Luna measured |
| LMArena Text: science | 1387 | 1409 | Inkling 2 Small |
| LMArena Text: medicine and healthcare | 1349 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: legal and government | 1392 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: business and finance | 1382 | 1415 | Inkling 2 Small |
| FACTS Benchmark Suite | 64.8% | Not measured | Only GPT-6 Luna measured |
| PRBench Legal | 42.9% | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 12 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| Design Arena: websites | 1268 | Not measured | Only GPT-6 Luna measured |
| Design Arena: UI components | 1291 | Not measured | Only GPT-6 Luna measured |
| Design Arena: game dev | 1253 | Not measured | Only GPT-6 Luna measured |
| Design Arena: data viz | 1208 | Not measured | Only GPT-6 Luna measured |
| Design Arena: 3D | 1194 | Not measured | Only GPT-6 Luna measured |
| Design Arena: all code categories | 1261 | Not measured | Only GPT-6 Luna measured |
| Design Arena: SVG | 1235 | Not measured | Only GPT-6 Luna measured |
| Design Arena: ASCII art | 1161 | Not measured | Only GPT-6 Luna measured |
| LiveBench design | 76.1% | Not measured | Only GPT-6 Luna measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index310
| Benchmark | GPT-6 Luna | Inkling 2 Small | Leader |
|---|---|---|---|
| Vals Index | 52.3% | 56.1% | Inkling 2 Small |
| LiveBench data analysis | 73.4% | Not measured | Only GPT-6 Luna measured |
| Public Benefits Bench (Vals AI) | 57.6% | Not measured | Only GPT-6 Luna measured |
| SAGE (Vals AI) | 83.4% | 88.6% | Inkling 2 Small |
| MedScribe (Vals AI) | 83.7% | Not measured | Only GPT-6 Luna measured |
| Long context reasoning (AA-LCR) | 83.3% | 83.3% | Tie |
| GDPval-AA | 1443 | 1630 | Inkling 2 Small |
| DTBench | 90.1% | Not measured | Only GPT-6 Luna measured |
| LMCA | 44.5% | Not measured | Only GPT-6 Luna measured |
| Furniture assembly | 44.2% | Not measured | Only GPT-6 Luna measured |
| LMArena Text | 1388 | 1403 | Inkling 2 Small |
| LMArena Text: instruction following | 1403 | 1408 | Inkling 2 Small |
| LMArena Text: creative writing | 1359 | 1348 | GPT-6 Luna |
| LMArena Vision | 1211 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: English | 1391 | 1418 | Inkling 2 Small |
| LMArena Text: non-English | 1380 | 1390 | Inkling 2 Small |
| LMArena Text: Chinese | 1422 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: German | 1384 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: Japanese | 1366 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: Korean | 1347 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: French | 1405 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: Spanish | 1395 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: Russian | 1388 | 1377 | GPT-6 Luna |
| LMArena Text: Polish | 1382 | Not measured | Only GPT-6 Luna measured |
| LMArena Text: multi-turn | 1393 | 1408 | Inkling 2 Small |
| LMArena Text: longer queries | 1403 | 1403 | Inkling 2 Small |
| LMArena Text: writing and language | 1382 | 1381 | GPT-6 Luna |
| LMArena Text: entertainment and media | 1359 | 1359 | Inkling 2 Small |
| LMArena Vision: English | 1204 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: Chinese | 1274 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: reading text | 1221 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: humor | 1168 | Not measured | Only GPT-6 Luna measured |
| LMArena Vision: creative writing | 1190 | Not measured | Only GPT-6 Luna measured |
| Neither model has a result on the other 16 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-6 Luna | Inkling 2 Small | 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$0.0003Fits in one request
Inkling 2 Small$0.0008Fits in one request
GPT-6 Luna costs less
Repository review
50K input + 3K output tokens
GPT-6 Luna$0.0065Fits in one request
Inkling 2 Small$0.018Fits in one request
GPT-6 Luna costs less
Agent loop
200K input + 10K output tokens
GPT-6 Luna$0.025Fits in one request
Inkling 2 Small$0.070Fits in one request
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-6 Luna | Thinking Machines Inkling 2 Small |
|---|---|---|
| Maker | OpenAI | Thinking Machines |
| Weights | No public weights listed | Open weights |
| Released | September 22, 2026 | November 9, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.10 | $0.30 |
| Output $/Mtok | $0.50 | $1.00 |
| Coding index | 54.9 | 48.9 |
| Independent measurements behind it | 5 | 3 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-6 Luna better than Inkling 2 Small for coding?
Pick GPT-6 Luna for coding: it scores 54.9 on the Vibe Coding Index against 48.9 for Inkling 2 Small. Pick Inkling 2 Small for work measured by LMArena WebDev or SWE-bench Multimodal (Vals AI), where it leads.
Which is cheaper, GPT-6 Luna or Inkling 2 Small?
GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens; Inkling 2 Small costs $0.30 and $1.00. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-6 Luna or Inkling 2 Small?
GPT-6 Luna, with 1.05M tokens against 1.05M.
How many benchmarks is the GPT-6 Luna vs Inkling 2 Small comparison based on?
48 benchmarks have a result for both models as of November 14, 2026. GPT-6 Luna's coding index rests on 5 independent benchmarks and Inkling 2 Small's on 3. Where only one model has a result, the page says so instead of naming a winner.