Model comparison
Llama 3-70B vs Qwen-14B
Qwen-14B is the stronger model overall, scoring 31.4 to 28.8 on the Noometry Index.
Last verified . 12 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Llama 3-70B scores higher in 5 categories and Qwen-14B in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen-14B leads 31.2 to 12.8.
Side by side
| Llama 3-70B | Qwen-14B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.8 | 31.4 |
| Released | 2024-04-18 | 2023-09-24 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 31 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 3-70B leads
Llama 3-70B: 35.8 (#218), Qwen-14B: 31.2 (#288)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Coding | 1206 | 1071 |
| BigCodeBench Instruct | 43.6% | — |
| BigCodeBench Complete | 54.5% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 69% | — |
Agentic & Tool Use Not comparable
Llama 3-70B: 21.1 (#139), Qwen-14B: —
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| Cybench | 5% | — |
Reasoning Qwen-14B leads
Llama 3-70B: 18.0 (#288), Qwen-14B: 19.6 (#257)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1195 | 1027 |
| Epoch Capabilities Index | 122.93 | 113.03 |
| Kagi LLM Benchmark | 35.1% | — |
| DTBench | 54.2% | — |
| BIG-Bench Hard | — | 55% |
| ForecastBench | 57.1 | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
| WinoGrande | 83.5% | — |
Math Qwen-14B leads
Llama 3-70B: 12.8 (#305), Qwen-14B: 31.2 (#227)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Math | 1218 | 1068 |
| OTIS Mock AIME 2024-2025 | 4.3% | — |
| MATH Level 5 | 22.6% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
Llama 3-70B: 20.8 (#277), Qwen-14B: —
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| MMLU | 79.3% | 66.3% |
| GPQA Diamond | 40.6% | — |
| LMArena Expert | 1149 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
Multilingual Llama 3-70B leads
Llama 3-70B: 33.6 (#251), Qwen-14B: 27.5 (#275)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1142 | 1041 |
| LMArena Chinese | 1114 | 1077 |
| LMArena French | 1232 | — |
| LMArena German | 1169 | — |
| LMArena Japanese | 1017 | — |
| LMArena Korean | 1017 | — |
| LMArena Russian | 1159 | — |
| LMArena Spanish | 1241 | — |
Instruction Following Llama 3-70B leads
Llama 3-70B: 62.5 (#238), Qwen-14B: 52.4 (#289)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1194 | 1031 |
Long Context Llama 3-70B leads
Llama 3-70B: 35.6 (#240), Qwen-14B: 31.3 (#280)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1174 | 1028 |
Writing & Preference Llama 3-70B leads
Llama 3-70B: 42.8 (#231), Qwen-14B: 27.6 (#299)
| Benchmark | Llama 3-70B | Qwen-14B |
|---|---|---|
| LMArena Text | 1221 | 1051 |
| LMArena Creative Writing | 1210 | 1028 |
| LMArena Multi-Turn | 1223 | 1022 |
Frequently asked questions
Is Llama 3-70B better than Qwen-14B?
Qwen-14B is the stronger model overall, scoring 31.4 to 28.8 on the Noometry Index.
Is Llama 3-70B or Qwen-14B better for coding?
Llama 3-70B scores higher on coding benchmarks: 35.8 versus 31.2 in the Noometry coding category.
How many benchmarks do Llama 3-70B and Qwen-14B share?
12 benchmarks have published results for both models. Llama 3-70B has 31 scored results on Noometry and Qwen-14B has 18.