# Claude Opus 4.7 vs GPT-5.4 mini

> Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 45.0 on the Noometry Index. GPT-5.4 mini costs 5.9× less per token, which makes it the better buy when Claude Opus 4.7's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-7-vs-gpt-5-4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 45

## Summary

- They share 45 benchmarks with published results for both. Claude Opus 4.7 scores higher in 10 categories and GPT-5.4 mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.7 leads 53.8 to 30.4.
- The biggest single-benchmark swing is ARC-AGI-2: 75.8% for Claude Opus 4.7 and 18.9% for GPT-5.4 mini.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $5 / $25 for Claude Opus 4.7.
- Claude Opus 4.7 accepts more context: 1M tokens versus 400K.

## Snapshot

| | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 58.3 | 45.0 |
| Rank | 19 | 76 |
| Context | 1M | 400K |
| Input $/M | $5 | $0.75 |
| Output $/M | $25 | $4.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4.7: 59.6 (#13)
- GPT-5.4 mini: 45.2 (#72)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| FrontierCode | 38.5% | 27% |
| LMArena WebDev | 1558 | 1397 |
| SciCode | 54.5% | 49.9% |
| WeirdML | 76.4% | 60.3% |
| LMArena Coding | 1518 | 1438 |
| ALE-Bench | 1,323 | 1,189 |
| SWE-bench Verified | 83.5% | — |
| GSO | 44.1% | — |
| MirrorCode | 31.1% | — |

## Agentic & Tool Use

- Claude Opus 4.7: 47.9 (#10)
- GPT-5.4 mini: 29.9 (#81)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| Terminal-Bench | 80.2% | — |
| APEX-Agents | 49.2% | — |
| OSWorld 2.0 | 18.2% | — |
| τ²-bench Banking | 40.2% | — |
| DeepResearch Bench | — | 36.3% |
| PostTrainBench | 28.6% | — |
| ExploitBench | 26.5% | — |
| GBAEval | 43.8% | — |
| GDP.pdf | 21% | — |
| LMArena Search | 1233 | — |
| Vending-Bench 2 | 10,937 | — |

## Reasoning

- Claude Opus 4.7: 53.8 (#29)
- GPT-5.4 mini: 30.4 (#85)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| ARC-AGI-2 | 75.8% | 18.9% |
| Kagi LLM Benchmark | 80.7% | 37.9% |
| NYT Connections (extended) | 39% | 61.8% |
| ARC-AGI-1 | 93.5% | 63.7% |
| CritPt | 12% | 10% |
| Chess Puzzles | 30% | 24% |
| Thematic Generalization | 72.8% | 61.7% |
| LMArena Hard Prompts | 1506 | 1424 |
| Mystery Game Puzzles | 28% | 11% |
| DTBench | 94.7% | 80% |
| LMCA | 52.2% | 40.8% |
| Epoch Capabilities Index | 156.25 | 148.84 |
| ForecastBench | 60.3 | 57 |
| SimpleBench | 61.7% | — |
| EBR-Bench | 19% | — |

## Math

- Claude Opus 4.7: 66.7 (#26)
- GPT-5.4 mini: 45.5 (#75)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 70.2% | 51.2% |
| FrontierMath Tier 4 | 31.7% | 9.8% |
| OTIS Mock AIME 2024-2025 | 97.8% | 88.9% |
| ProofBench | 54% | 21% |
| LMArena Math | 1499 | 1419 |
| FrontierMath (Feb 2025 set) | 43.8% | 28.3% |
| FrontierMath Tier 4 (v1) | 22.9% | 2.1% |
| MathArena Final-Answer Competitions | 73.6% | — |

## Knowledge

- Claude Opus 4.7: 62.6 (#23)
- GPT-5.4 mini: 51.5 (#67)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 90.2% | 86.9% |
| SimpleQA Verified | 51.7% | 29.4% |
| Vectara Hallucination Rate | 12% | 5.5% |
| LMArena Expert | 1521 | 1435 |
| Humanity's Last Exam | 36.2% | — |

## Multimodal

- Claude Opus 4.7: 41.2 (#38)
- GPT-5.4 mini: 39.7 (#56)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1316 | 1245 |
| Blueprint-Bench 2 | 24.5% | — |
| Furniture Assembly | 33.3% | — |
| LMArena Document | 1495 | — |

## Multilingual

- Claude Opus 4.7: 57.3 (#10)
- GPT-5.4 mini: 51.9 (#96)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1480 | 1405 |
| LMArena Chinese | 1531 | 1446 |
| LMArena French | 1503 | 1440 |
| LMArena German | 1495 | 1409 |
| LMArena Japanese | 1472 | 1374 |
| LMArena Korean | 1464 | 1368 |
| LMArena Russian | 1494 | 1417 |
| LMArena Spanish | 1495 | 1405 |

## Instruction Following

- Claude Opus 4.7: 78.4 (#10)
- GPT-5.4 mini: 74.1 (#102)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1498 | 1405 |

## Long Context

- Claude Opus 4.7: 46.2 (#25)
- GPT-5.4 mini: 43.0 (#112)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1505 | 1407 |

## Writing & Preference

- Claude Opus 4.7: 75.1 (#8)
- GPT-5.4 mini: 64.0 (#58)

| Benchmark | Claude Opus 4.7 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1490 | 1412 |
| LMArena Creative Writing | 1486 | 1370 |
| EQ-Bench Creative Writing | 1914 | 1665 |
| LMArena Multi-Turn | 1505 | 1429 |
| EQ-Bench 4 | 1311 | — |

## FAQ

### Is Claude Opus 4.7 better than GPT-5.4 mini?

Claude Opus 4.7 is the stronger model overall, scoring 58.3 to 45.0 on the Noometry Index. GPT-5.4 mini costs 5.9× less per token, which makes it the better buy when Claude Opus 4.7's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.7 or GPT-5.4 mini?

GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Claude Opus 4.7 lists at $5 and $25.

### Is Claude Opus 4.7 or GPT-5.4 mini better for coding?

Claude Opus 4.7 scores higher on coding benchmarks: 59.6 versus 45.2 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4.7 does, with 1M tokens against 400K.

### How many benchmarks do Claude Opus 4.7 and GPT-5.4 mini share?

45 benchmarks have published results for both models. Claude Opus 4.7 has 66 scored results on Noometry and GPT-5.4 mini has 46.