# Grok-3 mini vs Inkling

> Inkling is the stronger model overall, scoring 44.1 to 41.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-mini-vs-inkling
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Grok-3 mini scores higher in 3 categories and Inkling in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 13.6.
- The biggest single-benchmark swing is ARC-AGI-1: 16.5% for Grok-3 mini and 79.5% for Inkling.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | Grok-3 mini | Inkling |
|---|---|---|
| Provider | xAI | Thinking Machines Lab |
| Noometry Index | 41.2 | 44.1 |
| Rank | 141 | 80 |
| Context | — | 66K |
| Input $/M | — | $1.87 |
| Output $/M | — | $4.68 |
| Weights | Proprietary | Open |

## Coding

- Grok-3 mini: 40.8 (#131)
- Inkling: 34.5 (#234)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| WeirdML | 42.6% | 32.3% |
| LMArena Coding | 1379 | 1464 |
| FrontierCode | — | 14% |
| Aider Polyglot | 49.3% | — |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- Grok-3 mini: —
- Inkling: 29.6 (#85)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |

## Reasoning

- Grok-3 mini: 13.6 (#334)
- Inkling: 40.4 (#56)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| ARC-AGI-2 | 0.4% | 36.5% |
| ARC-AGI-1 | 16.5% | 79.5% |
| LMArena Hard Prompts | 1375 | 1451 |
| Epoch Capabilities Index | 140.35 | 148.54 |
| SimpleBench | — | 50% |
| Kagi LLM Benchmark | 61.3% | — |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |

## Math

- Grok-3 mini: 42.1 (#85)
- Inkling: 31.3 (#225)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 77.8% | 88.9% |
| LMArena Math | 1386 | 1479 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 31.8% | — |
| MATH Level 5 | 90.9% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |

## Knowledge

- Grok-3 mini: 46.4 (#81)
- Inkling: 55.1 (#49)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| GPQA Diamond | 76.3% | 88.3% |
| LMArena Expert | 1395 | 1465 |
| SimpleQA Verified | — | 40.3% |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |

## Multilingual

- Grok-3 mini: 48.1 (#145)
- Inkling: 54.0 (#52)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| LMArena Non-English | 1352 | 1434 |
| LMArena Chinese | 1387 | 1490 |
| LMArena French | 1357 | 1458 |
| LMArena German | 1349 | 1446 |
| LMArena Japanese | 1342 | 1429 |
| LMArena Korean | 1335 | 1404 |
| LMArena Russian | 1353 | 1429 |
| LMArena Spanish | 1381 | 1448 |

## Instruction Following

- Grok-3 mini: 78.5 (#9)
- Inkling: 75.1 (#71)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| LMArena Instruction Following | 1357 | 1426 |
| IFEval | 95.1% | — |

## Long Context

- Grok-3 mini: 41.0 (#147)
- Inkling: 43.8 (#86)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| LMArena Longer Query | 1372 | 1434 |
| Fiction.LiveBench | 66.7% | — |

## Writing & Preference

- Grok-3 mini: 52.5 (#169)
- Inkling: 65.2 (#51)

| Benchmark | Grok-3 mini | Inkling |
|---|---|---|
| LMArena Text | 1370 | 1441 |
| LMArena Creative Writing | 1342 | 1387 |
| LMArena Multi-Turn | 1355 | 1436 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | — | 1611 |
| WildBench | 65.1% | — |
| EQ-Bench 4 | — | 1226 |

## FAQ

### Is Grok-3 mini better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 41.2 on the Noometry Index.

### Is Grok-3 mini or Inkling better for coding?

Grok-3 mini scores higher on coding benchmarks: 40.8 versus 34.5 in the Noometry coding category.

### How many benchmarks do Grok-3 mini and Inkling share?

23 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Inkling has 41.