# GPT-5.3 Chat vs Inkling

> Inkling is the stronger model overall, scoring 44.1 to 42.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-5-3-chat-vs-inkling
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. GPT-5.3 Chat scores higher in 2 categories and Inkling in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Inkling leads 55.2 to 38.2.
- Inkling is cheaper at $1.87 / $4.68 per million input/output tokens, against $1.75 / $14 for GPT-5.3 Chat.
- GPT-5.3 Chat accepts more context: 128K tokens versus 66K.
- Inkling has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-5.3 Chat | Inkling |
|---|---|---|
| Provider | OpenAI | Thinking Machines Lab |
| Noometry Index | 42.1 | 44.1 |
| Rank | 109 | 75 |
| Context | 128K | 66K |
| Input $/M | $1.75 | $1.87 |
| Output $/M | $14 | $4.68 |
| Weights | Proprietary | Open |

## Coding

- GPT-5.3 Chat: 40.8 (#124)
- Inkling: 34.1 (#229)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Coding | 1408 | 1463 |
| FrontierCode | — | 14% |
| LMArena WebDev | — | 1412 |
| FrontierSWE | — | 4.1% |
| SciCode | — | 47% |
| WeirdML | — | 32.3% |
| ALE-Bench | — | 946 |

## Agentic & Tool Use

- GPT-5.3 Chat: —
- Inkling: 29.5 (#76)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| APEX-Agents | — | 33.8% |
| τ²-bench Banking | — | 25% |

## Reasoning

- GPT-5.3 Chat: 27.3 (#105)
- Inkling: 40.6 (#48)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Hard Prompts | 1399 | 1452 |
| ARC-AGI-2 | — | 36.5% |
| SimpleBench | — | 50% |
| ARC-AGI-1 | — | 79.5% |
| CritPt | — | 5.4% |
| Chess Puzzles | — | 21% |
| DTBench | — | 87.5% |
| LMCA | — | 37.6% |
| Epoch Capabilities Index | — | 148.6 |

## Math

- GPT-5.3 Chat: 35.4 (#146)
- Inkling: 31.2 (#198)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Math | 1389 | 1482 |
| FrontierMath (Tiers 1-3) | — | 33.3% |
| FrontierMath Tier 4 | — | 4.9% |
| OTIS Mock AIME 2024-2025 | — | 88.9% |
| ProofBench | — | 0% |

## Knowledge

- GPT-5.3 Chat: 38.2 (#137)
- Inkling: 55.2 (#44)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Expert | 1397 | 1468 |
| GPQA Diamond | — | 88.3% |
| SimpleQA Verified | — | 40.3% |

## Multilingual

- GPT-5.3 Chat: 50.5 (#119)
- Inkling: 54.2 (#49)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Non-English | 1383 | 1434 |
| LMArena Chinese | 1433 | 1495 |
| LMArena French | 1398 | 1453 |
| LMArena German | 1384 | 1456 |
| LMArena Japanese | 1351 | 1428 |
| LMArena Korean | 1347 | 1401 |
| LMArena Russian | 1400 | 1430 |
| LMArena Spanish | 1371 | 1442 |

## Instruction Following

- GPT-5.3 Chat: 73.0 (#124)
- Inkling: 75.3 (#65)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Instruction Following | 1378 | 1427 |

## Long Context

- GPT-5.3 Chat: 42.6 (#115)
- Inkling: 43.9 (#79)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Longer Query | 1396 | 1436 |

## Writing & Preference

- GPT-5.3 Chat: 63.3 (#65)
- Inkling: 65.5 (#48)

| Benchmark | GPT-5.3 Chat | Inkling |
|---|---|---|
| LMArena Text | 1389 | 1443 |
| LMArena Creative Writing | 1356 | 1390 |
| EQ-Bench Creative Writing | 1690 | 1611 |
| LMArena Multi-Turn | 1413 | 1436 |
| EQ-Bench 4 | — | 1226 |

## FAQ

### Is GPT-5.3 Chat better than Inkling?

Inkling is the stronger model overall, scoring 44.1 to 42.1 on the Noometry Index.

### Which is cheaper, GPT-5.3 Chat or Inkling?

Inkling is cheaper. It lists at $1.87 per million input tokens and $4.68 per million output tokens; GPT-5.3 Chat lists at $1.75 and $14.

### Is GPT-5.3 Chat or Inkling better for coding?

GPT-5.3 Chat scores higher on coding benchmarks: 40.8 versus 34.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-5.3 Chat does, with 128K tokens against 66K.

### How many benchmarks do GPT-5.3 Chat and Inkling share?

18 benchmarks have published results for both models. GPT-5.3 Chat has 18 scored results on Noometry and Inkling has 41.