# Claude Opus 4.5 vs Qwen3 14B

> Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 35.5 on the Noometry Index. Qwen3 14B costs 16× 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-qwen3-14b
- Last updated: 2026-10-12
- Shared benchmarks: 9

## Summary

- They share 9 benchmarks with published results for both. Claude Opus 4.5 scores higher in 5 categories and Qwen3 14B in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.5 leads 42.6 to 18.5.
- The biggest single-benchmark swing is Berkeley Function Calling Leaderboard: 77.5% for Claude Opus 4.5 and 41% for Qwen3 14B.
- Qwen3 14B is cheaper at $0.35 / $1.40 per million input/output tokens, against $5 / $25 for Claude Opus 4.5.
- Claude Opus 4.5 accepts more context: 200K tokens versus 131K.
- Qwen3 14B has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 50.5 | 35.5 |
| Rank | 47 | 225 |
| Context | 200K | 131K |
| Input $/M | $5 | $0.35 |
| Output $/M | $25 | $1.40 |
| Weights | Proprietary | Open |

## Coding

- Claude Opus 4.5: 54.8 (#27)
- Qwen3 14B: 37.3 (#195)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| SWE-bench Verified | 76.7% | — |
| SWE-bench Verified (bash only) | 76.8% | — |
| LMArena WebDev | 1494 | — |
| SWE-bench Multilingual | 70.7% | — |
| SciCode | — | 31.6% |
| GSO | 26.5% | — |
| WeirdML | 63.7% | — |
| LMArena Coding | 1504 | — |
| ALE-Bench | 1,025 | — |
| AlgoTune | 1.77 | — |

## Agentic & Tool Use

- Claude Opus 4.5: 47.3 (#12)
- Qwen3 14B: 29.6 (#83)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 77.5% | 41% |
| Terminal-Bench | 63.1% | — |
| 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% | — |
| Vending-Bench 2 | 4,967 | — |

## Reasoning

- Claude Opus 4.5: 42.6 (#51)
- Qwen3 14B: 18.5 (#280)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | 80.2% | 49.1% |
| Chess Puzzles | 12% | 4% |
| DTBench | 89.9% | 64% |
| LMCA | 44.5% | 18.2% |
| Epoch Capabilities Index | 150.09 | 138.23 |
| ARC-AGI-2 | 37.6% | — |
| SimpleBench | 62% | — |
| NYT Connections (extended) | 52.5% | — |
| ARC-AGI-1 | 80% | — |
| CritPt | — | 0% |
| EnigmaEval | 11.9% | — |
| EBR-Bench | 14.3% | — |
| LMArena Hard Prompts | 1476 | — |
| Mystery Game Puzzles | 22% | — |
| ForecastBench | 60.7 | — |

## Math

- Claude Opus 4.5: 38.6 (#132)
- Qwen3 14B: 38.6 (#133)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.1% | 66.4% |
| FrontierMath (Tiers 1-3) | 34.4% | — |
| FrontierMath Tier 4 | 4.9% | — |
| ProofBench | 36% | — |
| LMArena Math | 1463 | — |
| FrontierMath (Feb 2025 set) | 20.7% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Opus 4.5: 56.5 (#44)
- Qwen3 14B: 39.3 (#134)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| GPQA Diamond | 86% | 63.8% |
| Vectara Hallucination Rate | 10.9% | 5.4% |
| Humanity's Last Exam | 25.2% | — |
| SimpleQA Verified | 45.7% | — |
| LMArena Expert | 1487 | — |

## Multimodal

- Claude Opus 4.5: 31.4 (#107)
- Qwen3 14B: —

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| GeoBench | 75% | — |
| VPCT | 40% | — |
| Furniture Assembly | 28.3% | — |
| LMArena Document | 1462 | — |

## Multilingual

- Claude Opus 4.5: 54.3 (#47)
- Qwen3 14B: —

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1438 | — |
| LMArena Chinese | 1470 | — |
| LMArena French | 1471 | — |
| LMArena German | 1449 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1424 | — |
| LMArena Russian | 1447 | — |
| LMArena Spanish | 1458 | — |

## Instruction Following

- Claude Opus 4.5: 77.5 (#19)
- Qwen3 14B: —

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| LMArena Instruction Following | 1478 | — |

## Long Context

- Claude Opus 4.5: 46.5 (#22)
- Qwen3 14B: 38.1 (#204)

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
| CL-bench | 21.1% | — |
| LMArena Longer Query | 1480 | — |

## Writing & Preference

- Claude Opus 4.5: 68.1 (#28)
- Qwen3 14B: —

| Benchmark | Claude Opus 4.5 | Qwen3 14B |
|---|---|---|
| LMArena Text | 1451 | — |
| LMArena Creative Writing | 1445 | — |
| EQ-Bench Creative Writing | 1687 | — |
| LMArena Multi-Turn | 1466 | — |

## FAQ

### Is Claude Opus 4.5 better than Qwen3 14B?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 35.5 on the Noometry Index. Qwen3 14B costs 16× 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 Qwen3 14B?

Qwen3 14B is cheaper. It lists at $0.35 per million input tokens and $1.40 per million output tokens; Claude Opus 4.5 lists at $5 and $25.

### Is Claude Opus 4.5 or Qwen3 14B better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 54.8 versus 37.3 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4.5 does, with 200K tokens against 131K.

### How many benchmarks do Claude Opus 4.5 and Qwen3 14B share?

9 benchmarks have published results for both models. Claude Opus 4.5 has 69 scored results on Noometry and Qwen3 14B has 12.