# GPT-4.1 nano vs Inkling-Small

> Inkling-Small is the stronger model overall, scoring 46.5 to 27.7 on the Noometry Index. GPT-4.1 nano costs 3.6× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4-1-nano-vs-inkling-small
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GPT-4.1 nano scores higher in 0 categories and Inkling-Small in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling-Small leads 39.0 to 8.6.
- The biggest single-benchmark swing is ARC-AGI-1: 0% for GPT-4.1 nano and 84% for Inkling-Small.
- GPT-4.1 nano is cheaper at $0.10 / $0.40 per million input/output tokens, against $0.45 / $1.20 for Inkling-Small.
- GPT-4.1 nano accepts more context: 1.05M tokens versus 524K.
- Inkling-Small has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 27.7 | 46.5 |
| Rank | 322 | 57 |
| Context | 1.05M | 524K |
| Input $/M | $0.10 | $0.45 |
| Output $/M | $0.40 | $1.20 |
| Weights | Proprietary | Open |

## Coding

- GPT-4.1 nano: 23.8 (#325)
- Inkling-Small: 43.3 (#80)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| SciCode | 25.9% | 48.7% |
| LMArena Coding | 1306 | 1451 |
| Aider Polyglot | 8.9% | — |
| LMArena WebDev | — | 1408 |
| WeirdML | 19% | — |

## Agentic & Tool Use

- GPT-4.1 nano: 26.3 (#98)
- Inkling-Small: —

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| Berkeley Function Calling Leaderboard | 33% | — |

## Reasoning

- GPT-4.1 nano: 8.6 (#344)
- Inkling-Small: 39.0 (#53)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| ARC-AGI-2 | 0% | 40.1% |
| ARC-AGI-1 | 0% | 84% |
| CritPt | 0% | 8.3% |
| LMArena Hard Prompts | 1286 | 1424 |
| Epoch Capabilities Index | 129.62 | 150.13 |
| Kagi LLM Benchmark | 33.3% | — |
| Chess Puzzles | — | 18% |
| Mystery Game Puzzles | — | 6% |
| DTBench | 52.5% | — |
| LMCA | 5.5% | — |

## Math

- GPT-4.1 nano: 25.5 (#246)
- Inkling-Small: 44.9 (#65)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 28.9% | 90% |
| LMArena Math | 1274 | 1453 |
| FrontierMath (Tiers 1-3) | — | 46.3% |
| FrontierMath Tier 4 | — | 17.1% |
| ProofBench | — | 6% |
| Omni-MATH | 36.7% | — |
| MATH Level 5 | 70% | — |
| FrontierMath (Feb 2025 set) | 1% | — |

## Knowledge

- GPT-4.1 nano: 21.8 (#268)
- Inkling-Small: 48.3 (#71)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| GPQA Diamond | 48.9% | 88.5% |
| SimpleQA Verified | 6% | 19.1% |
| LMArena Expert | 1272 | 1442 |
| MMLU-Pro | 55% | — |
| GPQA (HELM) | 50.7% | — |

## Multimodal

- GPT-4.1 nano: 28.6 (#108)
- Inkling-Small: 38.4 (#59)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| LMArena Vision | 1063 | 1235 |

## Multilingual

- GPT-4.1 nano: 41.8 (#200)
- Inkling-Small: 51.7 (#102)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1260 | 1400 |
| LMArena Chinese | 1270 | 1458 |
| LMArena German | 1288 | 1412 |
| LMArena Japanese | 1199 | 1399 |
| LMArena Russian | 1261 | 1392 |
| LMArena French | — | 1438 |
| LMArena Korean | — | 1355 |
| LMArena Spanish | — | 1426 |

## Instruction Following

- GPT-4.1 nano: 68.0 (#188)
- Inkling-Small: 73.9 (#112)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1267 | 1397 |
| IFEval | 84.3% | — |

## Long Context

- GPT-4.1 nano: 23.7 (#291)
- Inkling-Small: 42.7 (#113)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1283 | 1401 |
| Fiction.LiveBench | 25% | — |

## Writing & Preference

- GPT-4.1 nano: 40.7 (#238)
- Inkling-Small: 59.8 (#102)

| Benchmark | GPT-4.1 nano | Inkling-Small |
|---|---|---|
| LMArena Text | 1285 | 1413 |
| LMArena Creative Writing | 1260 | 1330 |
| EQ-Bench Creative Writing | 946 | 1491 |
| LMArena Multi-Turn | 1277 | 1420 |
| WildBench | 81.2% | — |

## FAQ

### Is GPT-4.1 nano better than Inkling-Small?

Inkling-Small is the stronger model overall, scoring 46.5 to 27.7 on the Noometry Index. GPT-4.1 nano costs 3.6× less per token, which makes it the better buy when Inkling-Small's lead doesn't matter for your workload.

### Which is cheaper, GPT-4.1 nano or Inkling-Small?

GPT-4.1 nano is cheaper. It lists at $0.10 per million input tokens and $0.40 per million output tokens; Inkling-Small lists at $0.45 and $1.20.

### Is GPT-4.1 nano or Inkling-Small better for coding?

Inkling-Small scores higher on coding benchmarks: 43.3 versus 23.8 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 nano does, with 1.05M tokens against 524K.

### How many benchmarks do GPT-4.1 nano and Inkling-Small share?

24 benchmarks have published results for both models. GPT-4.1 nano has 38 scored results on Noometry and Inkling-Small has 33.