# Claude 3 Haiku vs QwQ-32B

> QwQ-32B is the stronger model overall, scoring 39.4 to 25.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-qwq-32b
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and QwQ-32B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where QwQ-32B leads 36.2 to 9.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 59.2% for QwQ-32B.
- QwQ-32B has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 25.7 | 39.4 |
| Rank | 333 | 153 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Open |

## Coding

- Claude 3 Haiku: 26.1 (#320)
- QwQ-32B: 34.9 (#220)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| BigCodeBench Instruct | 39.4% | 44.6% |
| LMArena Coding | 1199 | 1333 |
| BigCodeBench Complete | 50.1% | 54.4% |
| Aider Polyglot | — | 20.9% |
| WeirdML | 9.8% | — |
| LiveBench Coding | — | 72.2% |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Reasoning

- Claude 3 Haiku: 15.7 (#303)
- QwQ-32B: 22.9 (#171)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1326 |
| Epoch Capabilities Index | 118.3 | 137.6 |
| ForecastBench | 53.2 | 58.3 |
| Kagi LLM Benchmark | 34.2% | — |
| Chess Puzzles | — | 5% |
| LiveBench Reasoning | — | 83.5% |
| DTBench | 50.1% | — |
| LiveBench Data Analysis | — | 65% |
| LMCA | 8.8% | — |
| LiveBench | — | 72% |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.2 (#313)
- QwQ-32B: 36.2 (#123)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 59.2% |
| LMArena Math | 1188 | 1359 |
| LiveBench Math | — | 77.8% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#280)
- QwQ-32B: 37.2 (#150)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| GPQA Diamond | 36.3% | 65.3% |
| Confabulations | 34.2% | 15.6% |
| LMArena Expert | 1148 | 1323 |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.1 (#123)
- QwQ-32B: —

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.1 (#238)
- QwQ-32B: 45.0 (#171)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Non-English | 1178 | 1306 |
| LMArena Chinese | 1155 | 1378 |
| LMArena French | 1194 | 1337 |
| LMArena German | 1174 | 1313 |
| LMArena Japanese | 1102 | 1263 |
| LMArena Korean | 1108 | 1278 |
| LMArena Russian | 1204 | 1297 |
| LMArena Spanish | 1165 | 1353 |

## Instruction Following

- Claude 3 Haiku: 61.5 (#242)
- QwQ-32B: 72.8 (#131)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Instruction Following | 1174 | 1297 |
| LiveBench Instruction Following | — | 81.8% |

## Long Context

- Claude 3 Haiku: 36.1 (#232)
- QwQ-32B: 49.0 (#11)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Longer Query | 1191 | 1308 |
| Fiction.LiveBench | — | 83.3% |

## Writing & Preference

- Claude 3 Haiku: 29.8 (#286)
- QwQ-32B: 50.9 (#175)

| Benchmark | Claude 3 Haiku | QwQ-32B |
|---|---|---|
| LMArena Text | 1195 | 1329 |
| LMArena Creative Writing | 1157 | 1288 |
| EQ-Bench Creative Writing | 717 | 1257 |
| LMArena Multi-Turn | 1190 | 1314 |
| Short-Story Creative Writing | — | 80.2% |
| LiveBench Language | — | 51.4% |

## FAQ

### Is Claude 3 Haiku better than QwQ-32B?

QwQ-32B is the stronger model overall, scoring 39.4 to 25.7 on the Noometry Index.

### Is Claude 3 Haiku or QwQ-32B better for coding?

QwQ-32B scores higher on coding benchmarks: 34.9 versus 26.1 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and QwQ-32B share?

25 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and QwQ-32B has 36.