Model comparison
Llama 3-8B vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 31.4 to 25.5 on the Noometry Index.
Last verified . 13 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Llama 3-8B scores higher in 4 categories and Qwen-14B in 3 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen-14B leads 31.2 to 8.8.
Side by side
| Llama 3-8B | Qwen-14B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.5 | 31.4 |
| Released | 2024-04-18 | 2023-09-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 34 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Llama 3-8B: 31.0 (#289), Qwen-14B: 31.2 (#288)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Coding | 1152 | 1071 |
| BigCodeBench Instruct | 31.9% | — |
| BigCodeBench Complete | 36.9% | — |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Qwen-14B leads
Llama 3-8B: 14.3 (#326), Qwen-14B: 19.6 (#257)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1133 | 1027 |
| Epoch Capabilities Index | 116.45 | 113.03 |
| Chess Puzzles | 0% | — |
| DTBench | 43.9% | — |
| Adversarial NLI | 57.3% | — |
| BIG-Bench Hard | — | 55% |
| ForecastBench | 58.6 | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
| WinoGrande | 75.7% | — |
Math Qwen-14B leads
Llama 3-8B: 8.8 (#323), Qwen-14B: 31.2 (#227)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Math | 1151 | 1068 |
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| MATH Level 5 | 6.1% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Llama 3-8B: 7.8 (#308), Qwen-14B: —
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| ARC (AI2) Challenge | 82.8% | 84.4% |
| MMLU | 68.8% | 66.3% |
| GPQA Diamond | 26.1% | — |
| LMArena Expert | 1113 | — |
| BoolQ | — | 86.2% |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Llama 3-8B leads
Llama 3-8B: 30.8 (#261), Qwen-14B: 27.5 (#275)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1098 | 1041 |
| LMArena Chinese | 1076 | 1077 |
| LMArena French | 1159 | — |
| LMArena German | 1104 | — |
| LMArena Japanese | 967 | — |
| LMArena Korean | 1004 | — |
| LMArena Russian | 1109 | — |
| LMArena Spanish | 1173 | — |
Instruction Following Llama 3-8B leads
Llama 3-8B: 58.4 (#260), Qwen-14B: 52.4 (#289)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1127 | 1031 |
Long Context Llama 3-8B leads
Llama 3-8B: 34.2 (#251), Qwen-14B: 31.3 (#280)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1128 | 1028 |
Writing & Preference Llama 3-8B leads
Llama 3-8B: 37.5 (#256), Qwen-14B: 27.6 (#299)
| Benchmark | Llama 3-8B | Qwen-14B |
|---|---|---|
| LMArena Text | 1166 | 1051 |
| LMArena Creative Writing | 1150 | 1028 |
| LMArena Multi-Turn | 1152 | 1022 |
Frequently asked questions
Is Llama 3-8B better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 31.4 to 25.5 on the Noometry Index.
Is Llama 3-8B or Qwen-14B better for coding?
They score almost the same on coding (31.0 vs 31.2); test both on your own repository before choosing.
How many benchmarks do Llama 3-8B and Qwen-14B share?
13 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen-14B has 18.