# Claude Opus 4.5 vs GPT-5 Mini

> Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 41.8 on the Noometry Index. GPT-5 Mini costs 15× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-5-vs-gpt-5-mini
- Last updated: 2026-10-10
- Shared benchmarks: 48

## Summary

- They share 48 benchmarks with published results for both. Claude Opus 4.5 scores higher in 8 categories and GPT-5 Mini in 2 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.5 leads 42.6 to 23.9.
- The biggest single-benchmark swing is ARC-AGI-2: 37.6% for Claude Opus 4.5 and 4.4% for GPT-5 Mini.
- GPT-5 Mini is cheaper at $0.25 / $2 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- GPT-5 Mini accepts more context: 400K tokens versus 200K.

## Snapshot

| | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.5 | 41.8 |
| Rank | 47 | 128 |
| Context | 200K | 400K |
| Input $/M | $5 | $0.25 |
| Output $/M | $25 | $2 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4.5: 54.8 (#27)
- GPT-5 Mini: 40.1 (#146)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| SWE-bench Verified | 76.7% | 64.7% |
| SWE-bench Verified (bash only) | 76.8% | 59.8% |
| SWE-bench Multilingual | 70.7% | 39.7% |
| WeirdML | 63.7% | 52.7% |
| LMArena Coding | 1504 | 1406 |
| ALE-Bench | 1,025 | 799.77 |
| AlgoTune | 1.77 | 1.38 |
| LMArena WebDev | 1494 | — |
| SciCode | — | 39.2% |
| GSO | 26.5% | — |

## Agentic & Tool Use

- Claude Opus 4.5: 47.3 (#12)
- GPT-5 Mini: 31.1 (#70)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| Terminal-Bench | 63.1% | 34.8% |
| Berkeley Function Calling Leaderboard | 77.5% | 55.5% |
| Vending-Bench 2 | 4,967 | -31.18 |
| GDPval | 45.5% | — |
| Remote Labor Index | 3.8% | — |
| τ²-bench Airline | 84% | — |
| τ²-bench Banking | 24.7% | — |
| τ²-bench Retail | 79.6% | — |
| τ²-bench Telecom | 92.3% | — |
| Cybench | 82% | — |
| DeepResearch Bench | 54.8% | — |
| OSWorld | 66.3% | — |
| BALROG | 43.5% | — |
| LMArena Search | 1180 | — |
| METR Time Horizons | 75% | — |

## Reasoning

- Claude Opus 4.5: 42.6 (#51)
- GPT-5 Mini: 23.9 (#168)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| ARC-AGI-2 | 37.6% | 4.4% |
| Kagi LLM Benchmark | 80.2% | 70.3% |
| ARC-AGI-1 | 80% | 54.3% |
| Chess Puzzles | 12% | 30% |
| EnigmaEval | 11.9% | 8.2% |
| LMArena Hard Prompts | 1476 | 1380 |
| Mystery Game Puzzles | 22% | 10% |
| DTBench | 89.9% | 80.5% |
| LMCA | 44.5% | 34.2% |
| Epoch Capabilities Index | 150.09 | 145.52 |
| ForecastBench | 60.7 | 61 |
| SimpleBench | 62% | — |
| NYT Connections (extended) | 52.5% | — |
| CritPt | — | 0% |
| EBR-Bench | 14.3% | — |

## Math

- Claude Opus 4.5: 38.6 (#132)
- GPT-5 Mini: 46.7 (#69)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 34.4% | 46.7% |
| FrontierMath Tier 4 | 4.9% | 12.2% |
| OTIS Mock AIME 2024-2025 | 86.1% | 86.7% |
| ProofBench | 36% | 9% |
| LMArena Math | 1463 | 1378 |
| FrontierMath (Feb 2025 set) | 20.7% | 27.2% |
| FrontierMath Tier 4 (v1) | 4.2% | 6.3% |
| Omni-MATH | — | 72.2% |
| MATH Level 5 | — | 97.8% |

## Knowledge

- Claude Opus 4.5: 56.5 (#44)
- GPT-5 Mini: 45.6 (#86)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| GPQA Diamond | 86% | 75% |
| Humanity's Last Exam | 25.2% | 19.4% |
| SimpleQA Verified | 45.7% | 21.6% |
| Vectara Hallucination Rate | 10.9% | 12.9% |
| LMArena Expert | 1487 | 1379 |
| MMLU-Pro | — | 83.5% |
| Confabulations | — | 13.3% |
| GPQA (HELM) | — | 75.6% |

## Multimodal

- Claude Opus 4.5: 31.4 (#107)
- GPT-5 Mini: 35.6 (#85)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| VPCT | 40% | 40.2% |
| LMArena Vision | — | 1202 |
| GeoBench | 75% | — |
| Furniture Assembly | 28.3% | — |
| LMArena Document | 1462 | — |

## Multilingual

- Claude Opus 4.5: 54.3 (#47)
- GPT-5 Mini: 48.9 (#137)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| LMArena Non-English | 1438 | 1363 |
| LMArena Chinese | 1470 | 1385 |
| LMArena French | 1471 | 1386 |
| LMArena German | 1449 | 1366 |
| LMArena Japanese | 1416 | 1341 |
| LMArena Korean | 1424 | 1308 |
| LMArena Russian | 1447 | 1362 |
| LMArena Spanish | 1458 | 1355 |

## Instruction Following

- Claude Opus 4.5: 77.5 (#19)
- GPT-5 Mini: 76.2 (#46)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| LMArena Instruction Following | 1478 | 1357 |
| IFEval | — | 92.7% |

## Long Context

- Claude Opus 4.5: 46.5 (#22)
- GPT-5 Mini: 41.9 (#132)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| LMArena Longer Query | 1480 | 1355 |
| Fiction.LiveBench | — | 69.4% |
| CL-bench | 21.1% | — |

## Writing & Preference

- Claude Opus 4.5: 68.1 (#28)
- GPT-5 Mini: 55.2 (#148)

| Benchmark | Claude Opus 4.5 | GPT-5 Mini |
|---|---|---|
| LMArena Text | 1451 | 1373 |
| LMArena Creative Writing | 1445 | 1325 |
| EQ-Bench Creative Writing | 1687 | 1313 |
| LMArena Multi-Turn | 1466 | 1363 |
| Short-Story Creative Writing | — | 83.1% |
| WildBench | — | 85.5% |

## FAQ

### Is Claude Opus 4.5 better than GPT-5 Mini?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 41.8 on the Noometry Index. GPT-5 Mini costs 15× less per token, which makes it the better buy when Claude Opus 4.5's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4.5 or GPT-5 Mini?

GPT-5 Mini is cheaper. It lists at $0.25 per million input tokens and $2 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

### Is Claude Opus 4.5 or GPT-5 Mini better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 40.1 in the Noometry coding category.

### Which has the bigger context window?

GPT-5 Mini does, with 400K tokens against 200K.

### How many benchmarks do Claude Opus 4.5 and GPT-5 Mini share?

48 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and GPT-5 Mini has 60.