# GPT-6 Luna vs Inkling

> GPT-6 Luna is the stronger model overall, scoring 45.1 to 44.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-inkling
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. GPT-6 Luna scores higher in 2 categories and Inkling in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where GPT-6 Luna leads 54.7 to 34.1.
- The biggest single-benchmark swing is FrontierCode: 42.4% for GPT-6 Luna and 14% for Inkling.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- GPT-6 Luna accepts more context: 1.05M tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-6 Luna | Inkling |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 45.1 | 44.1 |
| Rank | 65 | 75 |
| Context | 1.05M | 66K |
| Input $/M | $0.10 | $1.87 |
| Output $/M | $0.50 | $4.68 |
| Weights | Proprietary | Open |

## Coding

- GPT-6 Luna: 54.7 (#24)
- Inkling: 34.1 (#229)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| FrontierCode | 42.4% | 14% |
| LMArena WebDev | 1583 | 1412 |
| LMArena Coding | 1448 | 1463 |
| DeepSWE | 66.6% | — |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- GPT-6 Luna: —
- Inkling: 29.5 (#76)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |

## Reasoning

- GPT-6 Luna: 31.7 (#68)
- Inkling: 40.6 (#48)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1415 | 1452 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| NYT Connections (extended) | 68.7% | — |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.6 |

## Math

- GPT-6 Luna: 35.6 (#140)
- Inkling: 31.2 (#198)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Math | 1396 | 1482 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |

## Knowledge

- GPT-6 Luna: 40.1 (#110)
- Inkling: 55.2 (#44)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Expert | 1458 | 1468 |
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |

## Multimodal

- GPT-6 Luna: 42.0 (#27)
- Inkling: —

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-6 Luna: 51.0 (#111)
- Inkling: 54.2 (#49)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Non-English | 1390 | 1434 |
| LMArena Chinese | 1425 | 1495 |
| LMArena Russian | 1392 | 1430 |
| LMArena Spanish | 1416 | 1442 |
| LMArena French | — | 1453 |
| LMArena German | — | 1456 |
| LMArena Japanese | — | 1428 |
| LMArena Korean | — | 1401 |

## Instruction Following

- GPT-6 Luna: 74.7 (#85)
- Inkling: 75.3 (#65)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Instruction Following | 1414 | 1427 |

## Long Context

- GPT-6 Luna: 43.3 (#95)
- Inkling: 43.9 (#79)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Longer Query | 1417 | 1436 |

## Writing & Preference

- GPT-6 Luna: 58.4 (#115)
- Inkling: 65.5 (#48)

| Benchmark | GPT-6 Luna | Inkling |
|---|---|---|
| LMArena Text | 1392 | 1443 |
| LMArena Creative Writing | 1357 | 1390 |
| LMArena Multi-Turn | 1400 | 1436 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |

## FAQ

### Is GPT-6 Luna better than Inkling?

GPT-6 Luna is the stronger model overall, scoring 45.1 to 44.1 on the Noometry Index.

### Which is cheaper, GPT-6 Luna or Inkling?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Inkling lists at $1.87 and $4.68.

### Is GPT-6 Luna or Inkling better for coding?

GPT-6 Luna scores higher on coding benchmarks: 54.7 versus 34.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 66K.

### How many benchmarks do GPT-6 Luna and Inkling share?

15 benchmarks have published results for both models. GPT-6 Luna has 18 scored results on Noometry and Inkling has 41.