# Claude 3 Haiku vs Qwen3.8 Max

> Qwen3.8 Max is the stronger model overall, scoring 56.8 to 25.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-haiku-vs-qwen3-8-max
- Last updated: 2026-10-12
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Qwen3.8 Max in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen3.8 Max leads 73.2 to 9.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 1.8% for Claude 3 Haiku and 100% for Qwen3.8 Max.

## Snapshot

| | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 25.9 | 56.8 |
| Rank | 340 | 22 |
| Context | — | 1M |
| Input $/M | — | $2 |
| Output $/M | — | $6 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Haiku: 26.4 (#325)
- Qwen3.8 Max: 53.5 (#29)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Coding | 1199 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| SciCode | — | 53.2% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |

## Agentic & Tool Use

- Claude 3 Haiku: —
- Qwen3.8 Max: 45.4 (#14)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |

## Reasoning

- Claude 3 Haiku: 16.3 (#307)
- Qwen3.8 Max: 54.4 (#26)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1496 |
| DTBench | 50.1% | 92% |
| LMCA | 8.8% | 46.2% |
| Epoch Capabilities Index | 118.35 | 156.41 |
| Kagi LLM Benchmark | 34.2% | — |
| NYT Connections (extended) | — | 88.3% |
| CritPt | — | 20% |
| Chess Puzzles | — | 40% |
| Mystery Game Puzzles | — | 38% |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |

## Math

- Claude 3 Haiku: 9.8 (#319)
- Qwen3.8 Max: 73.2 (#20)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.8% | 100% |
| LMArena Math | 1188 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| MATH Level 5 | 14.9% | — |

## Knowledge

- Claude 3 Haiku: 17.3 (#285)
- Qwen3.8 Max: 61.7 (#27)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 36.3% | 92.7% |
| LMArena Expert | 1148 | 1507 |
| SimpleQA Verified | — | 47.3% |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |

## Multimodal

- Claude 3 Haiku: 23.6 (#128)
- Qwen3.8 Max: 37.2 (#75)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | 950 | 1314 |
| Furniture Assembly | — | 20% |
| ScienceQA | 72% | — |

## Multilingual

- Claude 3 Haiku: 36.0 (#243)
- Qwen3.8 Max: 56.7 (#18)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1178 | 1472 |
| LMArena Chinese | 1155 | 1538 |
| LMArena French | 1195 | 1503 |
| LMArena German | 1174 | 1483 |
| LMArena Japanese | 1102 | 1467 |
| LMArena Korean | 1109 | 1461 |
| LMArena Russian | 1204 | 1481 |
| LMArena Spanish | 1166 | 1492 |

## Instruction Following

- Claude 3 Haiku: 61.3 (#247)
- Qwen3.8 Max: 77.6 (#17)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1173 | 1479 |

## Long Context

- Claude 3 Haiku: 36.1 (#237)
- Qwen3.8 Max: 45.6 (#31)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1190 | 1489 |

## Writing & Preference

- Claude 3 Haiku: 29.7 (#291)
- Qwen3.8 Max: 67.1 (#30)

| Benchmark | Claude 3 Haiku | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1195 | 1483 |
| LMArena Creative Writing | 1157 | 1479 |
| LMArena Multi-Turn | 1190 | 1489 |
| EQ-Bench Creative Writing | 717 | — |

## FAQ

### Is Claude 3 Haiku better than Qwen3.8 Max?

Qwen3.8 Max is the stronger model overall, scoring 56.8 to 25.9 on the Noometry Index.

### Is Claude 3 Haiku or Qwen3.8 Max better for coding?

Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 26.4 in the Noometry coding category.

### How many benchmarks do Claude 3 Haiku and Qwen3.8 Max share?

23 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Qwen3.8 Max has 39.