# GPT-4 Turbo vs Inkling-Small

> Inkling-Small is the stronger model overall, scoring 46.5 to 30.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-turbo-vs-inkling-small
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GPT-4 Turbo scores higher in 0 categories and Inkling-Small in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Inkling-Small leads 44.9 to 8.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.7% for GPT-4 Turbo and 90% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $10 / $30 for GPT-4 Turbo.
- Inkling-Small accepts more context: 524K tokens versus 128K.
- Inkling-Small has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 30.3 | 46.5 |
| Rank | 281 | 57 |
| Context | 128K | 524K |
| Input $/M | $10 | $0.45 |
| Output $/M | $30 | $1.20 |
| Weights | Proprietary | Open |

## Coding

- GPT-4 Turbo: 33.4 (#247)
- Inkling-Small: 43.3 (#80)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Coding | 1268 | 1451 |
| LMArena WebDev | — | 1408 |
| SciCode | — | 48.7% |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |

## Agentic & Tool Use

- GPT-4 Turbo: —
- Inkling-Small: —

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| METR Time Horizons | 36.7% | — |

## Reasoning

- GPT-4 Turbo: 14.9 (#311)
- Inkling-Small: 39.0 (#53)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| Chess Puzzles | 6% | 18% |
| LMArena Hard Prompts | 1251 | 1424 |
| Epoch Capabilities Index | 127.25 | 150.13 |
| ARC-AGI-2 | — | 40.1% |
| SimpleBench | 25.1% | — |
| ARC-AGI-1 | — | 84% |
| CritPt | — | 8.3% |
| Mystery Game Puzzles | — | 6% |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| ForecastBench | 59.4 | — |

## Math

- GPT-4 Turbo: 8.6 (#316)
- Inkling-Small: 44.9 (#65)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 46.3% |
| OTIS Mock AIME 2024-2025 | 6.7% | 90% |
| LMArena Math | 1272 | 1453 |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| MATH Level 5 | 46.7% | — |

## Knowledge

- GPT-4 Turbo: 24.3 (#263)
- Inkling-Small: 48.3 (#71)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| GPQA Diamond | 46.6% | 88.5% |
| LMArena Expert | 1223 | 1442 |
| SimpleQA Verified | — | 19.1% |
| Confabulations | 28.4% | — |
| MMLU | 81.3% | — |

## Multimodal

- GPT-4 Turbo: 30.0 (#104)
- Inkling-Small: 38.4 (#59)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Vision | 1090 | 1235 |

## Multilingual

- GPT-4 Turbo: 40.7 (#211)
- Inkling-Small: 51.7 (#102)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1245 | 1400 |
| LMArena Chinese | 1242 | 1458 |
| LMArena French | 1275 | 1438 |
| LMArena German | 1259 | 1412 |
| LMArena Japanese | 1194 | 1399 |
| LMArena Korean | 1187 | 1355 |
| LMArena Russian | 1259 | 1392 |
| LMArena Spanish | 1260 | 1426 |

## Instruction Following

- GPT-4 Turbo: 66.0 (#211)
- Inkling-Small: 73.9 (#112)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1250 | 1397 |

## Long Context

- GPT-4 Turbo: 38.0 (#200)
- Inkling-Small: 42.7 (#113)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1254 | 1401 |

## Writing & Preference

- GPT-4 Turbo: 47.9 (#201)
- Inkling-Small: 59.8 (#102)

| Benchmark | GPT-4 Turbo | Inkling-Small |
|---|---|---|
| LMArena Text | 1272 | 1413 |
| LMArena Creative Writing | 1269 | 1330 |
| LMArena Multi-Turn | 1267 | 1420 |
| EQ-Bench Creative Writing | — | 1491 |

## FAQ

### Is GPT-4 Turbo better than Inkling-Small?

Inkling-Small is the stronger model overall, scoring 46.5 to 30.3 on the Noometry Index.

### Which is cheaper, GPT-4 Turbo or Inkling-Small?

Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; GPT-4 Turbo lists at $10 and $30.

### Is GPT-4 Turbo or Inkling-Small better for coding?

Inkling-Small scores higher on coding benchmarks: 43.3 versus 33.4 in the Noometry coding category.

### Which has the bigger context window?

Inkling-Small does, with 524K tokens against 128K.

### How many benchmarks do GPT-4 Turbo and Inkling-Small share?

23 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Inkling-Small has 33.