# GPT-4o mini vs Inkling

> Inkling is the stronger model overall, scoring 44.1 to 25.4 on the Noometry Index. GPT-4o mini costs 9.8× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-4o-mini-vs-inkling
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. GPT-4o mini scores higher in 0 categories and Inkling in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.2 to 17.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.9% for GPT-4o mini and 88.9% for Inkling.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- GPT-4o mini accepts more context: 128K tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4o mini | Inkling |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 25.4 | 44.1 |
| Rank | 337 | 75 |
| Context | 128K | 66K |
| Input $/M | $0.15 | $1.87 |
| Output $/M | $0.60 | $4.68 |
| Weights | Proprietary | Open |

## Coding

- GPT-4o mini: 21.6 (#330)
- Inkling: 34.1 (#229)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| WeirdML | 11.8% | 32.3% |
| LMArena Coding | 1290 | 1463 |
| FrontierCode | — | 14% |
| Aider Polyglot | 3.6% | — |
| LMArena WebDev | — | 1412 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| ALE-Bench | — | 946 |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |

## Agentic & Tool Use

- GPT-4o mini: 27.2 (#95)
- Inkling: 29.5 (#76)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |
| BALROG | 17.4% | — |

## Reasoning

- GPT-4o mini: 8.7 (#343)
- Inkling: 40.6 (#48)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| ARC-AGI-2 | 0% | 36.5% |
| SimpleBench | 10.7% | 50% |
| Chess Puzzles | 0% | 21% |
| LMArena Hard Prompts | 1267 | 1452 |
| DTBench | 54.4% | 87.5% |
| LMCA | 10.4% | 37.6% |
| Epoch Capabilities Index | 126.56 | 148.6 |
| Kagi LLM Benchmark | 28.8% | — |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| LiveBench Data Analysis | 50% | — |
| LiveBench | 41.3% | — |
| PIQA | 88.7% | — |

## Math

- GPT-4o mini: 9.9 (#307)
- Inkling: 31.2 (#198)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.7% | 33.3% |
| OTIS Mock AIME 2024-2025 | 6.9% | 88.9% |
| LMArena Math | 1267 | 1482 |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 0% |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| MATH Level 5 | 52.6% | — |
| GSM8K | 91.3% | — |

## Knowledge

- GPT-4o mini: 17.7 (#279)
- Inkling: 55.2 (#44)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| GPQA Diamond | 37.7% | 88.3% |
| SimpleQA Verified | 8.3% | 40.3% |
| LMArena Expert | 1235 | 1468 |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| GPQA (HELM) | 36.8% | — |
| BoolQ | 88.7% | — |
| MMLU | 81.8% | — |

## Multimodal

- GPT-4o mini: 25.3 (#117)
- Inkling: —

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| LMArena Vision | 1066 | — |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |

## Multilingual

- GPT-4o mini: 42.2 (#194)
- Inkling: 54.2 (#49)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| LMArena Non-English | 1266 | 1434 |
| LMArena Chinese | 1264 | 1495 |
| LMArena French | 1296 | 1453 |
| LMArena German | 1272 | 1456 |
| LMArena Japanese | 1216 | 1428 |
| LMArena Korean | 1195 | 1401 |
| LMArena Russian | 1275 | 1430 |
| LMArena Spanish | 1276 | 1442 |

## Instruction Following

- GPT-4o mini: 62.1 (#234)
- Inkling: 75.3 (#65)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| LMArena Instruction Following | 1259 | 1427 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |

## Long Context

- GPT-4o mini: 39.1 (#181)
- Inkling: 43.9 (#79)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| LMArena Longer Query | 1289 | 1436 |

## Writing & Preference

- GPT-4o mini: 39.7 (#243)
- Inkling: 65.5 (#48)

| Benchmark | GPT-4o mini | Inkling |
|---|---|---|
| LMArena Text | 1286 | 1443 |
| LMArena Creative Writing | 1268 | 1390 |
| EQ-Bench Creative Writing | 873 | 1611 |
| LMArena Multi-Turn | 1285 | 1436 |
| Short-Story Creative Writing | 67.2% | — |
| WildBench | 79.1% | — |
| EQ-Bench 4 | — | 1226 |
| LiveBench Language | 28.6% | — |

## FAQ

### Is GPT-4o mini better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 25.4 on the Noometry Index. GPT-4o mini costs 9.8× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.

### Which is cheaper, GPT-4o mini or Inkling?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Inkling lists at $1.87 and $4.68.

### Is GPT-4o mini or Inkling better for coding?

Inkling scores higher on coding benchmarks: 34.1 versus 21.6 in the Noometry coding category.

### Which has the bigger context window?

GPT-4o mini does, with 128K tokens against 66K.

### How many benchmarks do GPT-4o mini and Inkling share?

29 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Inkling has 41.