Model comparison
Claude 3 Haiku vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 30.8 to 25.7 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Claude 3 Haiku scores higher in 4 categories and Qwen-14B in 3 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen-14B leads 28.7 to 9.2.
- Qwen-14B has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Haiku | Qwen-14B | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 25.7 | 30.8 |
| Released | 2024-03-07 | 2023-09-24 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 37 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen-14B leads
Claude 3 Haiku: 26.1 (#320), Qwen-14B: 30.6 (#283)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Coding | 1199 | 1072 |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Reasoning Qwen-14B leads
Claude 3 Haiku: 15.7 (#303), Qwen-14B: 18.7 (#255)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1027 |
| Epoch Capabilities Index | 118.3 | 112.8 |
| Kagi LLM Benchmark | 34.2% | — |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| BIG-Bench Hard | — | 55% |
| ForecastBench | 53.2 | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
| WinoGrande | 74.2% | — |
Math Qwen-14B leads
Claude 3 Haiku: 9.2 (#313), Qwen-14B: 28.7 (#224)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Math | 1188 | 1068 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| MATH Level 5 | 14.9% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Claude 3 Haiku: 17.3 (#280), Qwen-14B: —
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| MMLU | 73.8% | 66.3% |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| LMArena Expert | 1148 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
Multimodal Not comparable
Claude 3 Haiku: 23.1 (#123), Qwen-14B: —
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Vision | 950 | — |
| ScienceQA | 72% | — |
Multilingual Claude 3 Haiku leads
Claude 3 Haiku: 36.1 (#238), Qwen-14B: 27.7 (#270)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1178 | 1042 |
| LMArena Chinese | 1155 | 1077 |
| LMArena French | 1194 | — |
| LMArena German | 1174 | — |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1108 | — |
| LMArena Russian | 1204 | — |
| LMArena Spanish | 1165 | — |
Instruction Following Claude 3 Haiku leads
Claude 3 Haiku: 61.5 (#242), Qwen-14B: 52.6 (#284)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1174 | 1031 |
Long Context Claude 3 Haiku leads
Claude 3 Haiku: 36.1 (#232), Qwen-14B: 31.3 (#276)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1191 | 1029 |
Writing & Preference Claude 3 Haiku leads
Claude 3 Haiku: 29.8 (#286), Qwen-14B: 27.8 (#294)
| Benchmark | Claude 3 Haiku | Qwen-14B |
|---|---|---|
| LMArena Text | 1195 | 1051 |
| LMArena Creative Writing | 1157 | 1028 |
| LMArena Multi-Turn | 1190 | 1022 |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 30.8 to 25.7 on the Noometry Index.
Is Claude 3 Haiku or Qwen-14B better for coding?
Qwen-14B scores higher on coding benchmarks: 30.6 versus 26.1 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Qwen-14B share?
12 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Qwen-14B has 18.