Model comparison
Llama 13b vs Qwen1.5-14B
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 24.4 on the Noometry Index.
Last verified . 9 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 9 benchmarks with published results for both. Llama 13b scores higher in 0 categories and Qwen1.5-14B in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Qwen1.5-14B leads 56.8 to 36.7.
Side by side
| Llama 13b | Qwen1.5-14B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 24.4 | 32.7 |
| Released | 2023-02-24 | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-14B leads
Llama 13b: 21.4 (#337), Qwen1.5-14B: 33.1 (#263)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 683 | 1138 |
Reasoning Qwen1.5-14B leads
Llama 13b: 14.0 (#329), Qwen1.5-14B: 21.4 (#223)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 728 | 1113 |
| BIG-Bench Hard | 37.9% | — |
| Epoch Capabilities Index | 100.58 | — |
| HellaSwag | 79.2% | — |
| LAMBADA | 75.2% | — |
| PIQA | 80.1% | — |
| WinoGrande | 73% | — |
Math Qwen1.5-14B leads
Llama 13b: 26.7 (#256), Qwen1.5-14B: 32.4 (#215)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 838 | 1125 |
| GSM8K | 20.6% | — |
Knowledge Not comparable
Llama 13b: —, Qwen1.5-14B: 29.8 (#232)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| MMLU | 47.7% | 68.6% |
| LMArena Expert | — | 1094 |
| ARC (AI2) Challenge | 52.7% | — |
| BoolQ | 78.7% | — |
| OpenBookQA | 56.4% | — |
| TriviaQA | 77.9% | — |
Multimodal Not comparable
Llama 13b: —, Qwen1.5-14B: —
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| ScienceQA | 43.3% | — |
Multilingual Qwen1.5-14B leads
Llama 13b: 16.6 (#297), Qwen1.5-14B: 30.7 (#262)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 819 | 1095 |
| LMArena Chinese | — | 1147 |
| LMArena French | — | 1116 |
| LMArena German | — | 1043 |
| LMArena Japanese | — | 1019 |
| LMArena Russian | — | 1046 |
| LMArena Spanish | — | 1085 |
Instruction Following Qwen1.5-14B leads
Llama 13b: 36.7 (#305), Qwen1.5-14B: 56.8 (#271)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 781 | 1102 |
Long Context Not comparable
Llama 13b: —, Qwen1.5-14B: 33.7 (#257)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | — | 1113 |
Writing & Preference Qwen1.5-14B leads
Llama 13b: 13.8 (#312), Qwen1.5-14B: 33.6 (#276)
| Benchmark | Llama 13b | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 834 | 1128 |
| LMArena Creative Writing | 794 | 1091 |
| LMArena Multi-Turn | 753 | 1110 |
Frequently asked questions
Is Llama 13b better than Qwen1.5-14B?
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 24.4 on the Noometry Index.
Is Llama 13b or Qwen1.5-14B better for coding?
Qwen1.5-14B scores higher on coding benchmarks: 33.1 versus 21.4 in the Noometry coding category.
How many benchmarks do Llama 13b and Qwen1.5-14B share?
9 benchmarks have published results for both models. Llama 13b has 21 scored results on Noometry and Qwen1.5-14B has 17.