# Inkling-Small vs Step 3

> Inkling-Small is the stronger model overall, scoring 46.5 to 40.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/inkling-small-vs-step-3
- Last updated: 2026-10-11
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Inkling-Small scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling-Small leads 48.2 to 36.8.

## Snapshot

| | Inkling-Small | Step 3 |
|---|---|---|
| Provider | Thinking Machines Lab | StepFun |
| Noometry Index | 46.5 | 40.5 |
| Rank | 63 | 149 |
| Context | 524K | — |
| Input $/M | $0.45 | — |
| Output $/M | $1.20 | — |
| Weights | Open | Open |

## Coding

- Inkling-Small: 43.6 (#85)
- Step 3: 40.1 (#147)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Coding | 1451 | 1367 |
| LMArena WebDev | 1409 | — |
| SciCode | 48.7% | — |

## Reasoning

- Inkling-Small: 38.6 (#63)
- Step 3: 28.4 (#105)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1423 | 1355 |
| ARC-AGI-2 | 40.1% | — |
| Kagi LLM Benchmark | — | 62.3% |
| ARC-AGI-1 | 84% | — |
| CritPt | 8.3% | — |
| Chess Puzzles | 18% | — |
| Mystery Game Puzzles | 6% | — |
| Epoch Capabilities Index | 150.15 | — |

## Math

- Inkling-Small: 45.1 (#77)
- Step 3: 37.6 (#148)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Math | 1459 | 1366 |
| FrontierMath (Tiers 1-3) | 46.3% | — |
| FrontierMath Tier 4 | 17.1% | — |
| OTIS Mock AIME 2024-2025 | 90% | — |
| ProofBench | 6% | — |

## Knowledge

- Inkling-Small: 48.2 (#77)
- Step 3: 36.8 (#164)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Expert | 1442 | 1333 |
| GPQA Diamond | 88.5% | — |
| SimpleQA Verified | 19.1% | — |

## Multimodal

- Inkling-Small: 39.1 (#62)
- Step 3: 35.5 (#86)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Vision | 1235 | 1177 |

## Multilingual

- Inkling-Small: 51.7 (#104)
- Step 3: 46.3 (#159)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Non-English | 1402 | 1327 |
| LMArena Chinese | 1465 | 1397 |
| LMArena German | 1405 | 1371 |
| LMArena Korean | 1363 | 1269 |
| LMArena Russian | 1391 | 1331 |
| LMArena Spanish | 1428 | 1371 |
| LMArena French | 1436 | — |
| LMArena Japanese | 1405 | — |

## Instruction Following

- Inkling-Small: 73.8 (#114)
- Step 3: 70.4 (#164)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1399 | 1332 |

## Long Context

- Inkling-Small: 42.7 (#118)
- Step 3: 40.3 (#157)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Longer Query | 1401 | 1326 |

## Writing & Preference

- Inkling-Small: 59.6 (#107)
- Step 3: 54.3 (#151)

| Benchmark | Inkling-Small | Step 3 |
|---|---|---|
| LMArena Text | 1414 | 1350 |
| LMArena Creative Writing | 1331 | 1321 |
| LMArena Multi-Turn | 1418 | 1341 |
| EQ-Bench Creative Writing | 1491 | — |

## FAQ

### Is Inkling-Small better than Step 3?

Inkling-Small is the stronger model overall, scoring 46.5 to 40.5 on the Noometry Index.

### Is Inkling-Small or Step 3 better for coding?

Inkling-Small scores higher on coding benchmarks: 43.6 versus 40.1 in the Noometry coding category.

### How many benchmarks do Inkling-Small and Step 3 share?

16 benchmarks have published results for both models. Inkling-Small has 33 scored results on Noometry and Step 3 has 17.