# GPT-5.4 Pro vs Inkling

> GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 44.1 on the Noometry Index. Inkling costs 26× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-4-pro-vs-inkling
- Last updated: 2026-10-11
- Shared benchmarks: 11

## Summary

- They share 11 benchmarks with published results for both. GPT-5.4 Pro scores higher in 4 categories and Inkling in 0 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.4 Pro leads 72.4 to 31.3.
- The biggest single-benchmark swing is FrontierMath Tier 4: 58.5% for GPT-5.4 Pro and 4.9% for Inkling.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $30 / $180 for GPT-5.4 Pro.
- GPT-5.4 Pro accepts more context: 1.05M tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.4 Pro | Inkling |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 58.9 | 44.1 |
| Rank | 18 | 80 |
| Context | 1.05M | 66K |
| Input $/M | $30 | $1.87 |
| Output $/M | $180 | $4.68 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.4 Pro: 43.2 (#90)
- Inkling: 34.5 (#234)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| WeirdML | 57.4% | 32.3% |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1413 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| LMArena Coding | — | 1464 |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- GPT-5.4 Pro: —
- Inkling: 29.6 (#85)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |

## Reasoning

- GPT-5.4 Pro: 70.7 (#13)
- Inkling: 40.4 (#56)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| ARC-AGI-2 | 83.3% | 36.5% |
| SimpleBench | 74.1% | 50% |
| ARC-AGI-1 | 94.5% | 79.5% |
| CritPt | 30% | 5.4% |
| Chess Puzzles | 58.6% | 21% |
| Epoch Capabilities Index | 158.93 | 148.54 |
| EnigmaEval | 23.8% | — |
| LMArena Hard Prompts | — | 1451 |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |

## Math

- GPT-5.4 Pro: 72.4 (#22)
- Inkling: 31.3 (#225)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| FrontierMath (Tiers 1-3) | 82.5% | 33.3% |
| FrontierMath Tier 4 | 58.5% | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |
| LMArena Math | — | 1479 |
| FrontierMath (Feb 2025 set) | 50% | — |
| FrontierMath Tier 4 (v1) | 37.5% | — |

## Knowledge

- GPT-5.4 Pro: 68.3 (#7)
- Inkling: 55.1 (#49)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| GPQA Diamond | 94.6% | 88.3% |
| SimpleQA Verified | 46.3% | 40.3% |
| Humanity's Last Exam | 44.3% | — |
| Vectara Hallucination Rate | 8.3% | — |
| LMArena Expert | — | 1465 |

## Multilingual

- GPT-5.4 Pro: —
- Inkling: 54.0 (#52)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| LMArena Non-English | — | 1434 |
| LMArena Chinese | — | 1490 |
| LMArena French | — | 1458 |
| LMArena German | — | 1446 |
| LMArena Japanese | — | 1429 |
| LMArena Korean | — | 1404 |
| LMArena Russian | — | 1429 |
| LMArena Spanish | — | 1448 |

## Instruction Following

- GPT-5.4 Pro: —
- Inkling: 75.1 (#71)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| LMArena Instruction Following | — | 1426 |

## Long Context

- GPT-5.4 Pro: —
- Inkling: 43.8 (#86)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| LMArena Longer Query | — | 1434 |

## Writing & Preference

- GPT-5.4 Pro: —
- Inkling: 65.2 (#51)

| Benchmark | GPT-5.4 Pro | Inkling |
|---|---|---|
| LMArena Text | — | 1441 |
| LMArena Creative Writing | — | 1387 |
| EQ-Bench Creative Writing | — | 1611 |
| EQ-Bench 4 | — | 1226 |
| LMArena Multi-Turn | — | 1436 |

## FAQ

### Is GPT-5.4 Pro better than Inkling?

GPT-5.4 Pro is the stronger model overall, scoring 58.9 to 44.1 on the Noometry Index. Inkling costs 26× less per token, which makes it the better buy when GPT-5.4 Pro's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.4 Pro or Inkling?

Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; GPT-5.4 Pro lists at $30 and $180.

### Is GPT-5.4 Pro or Inkling better for coding?

GPT-5.4 Pro scores higher on coding benchmarks: 43.2 versus 34.5 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.4 Pro does, with 1.05M tokens against 66K.

### How many benchmarks do GPT-5.4 Pro and Inkling share?

11 benchmarks have published results for both models. GPT-5.4 Pro has 16 scored results on Noometry and Inkling has 41.