Model comparison
Llama 3.1-8B vs Qwen2-72B
Qwen2-72B is the stronger model overall, scoring 30.0 to 23.0 on the Noometry Index.
Last verified . 24 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. Llama 3.1-8B scores higher in 1 category and Qwen2-72B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2-72B leads 30.2 to 10.2.
- The biggest single-benchmark swing is MATH Level 5: 22.9% for Llama 3.1-8B and 39.1% for Qwen2-72B.
Side by side
| Llama 3.1-8B | Qwen2-72B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 23.0 | 30.0 |
| Released | 2024-07-23 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.08 | — |
| Results tracked | 43 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2-72B leads
Llama 3.1-8B: 20.2 (#340), Qwen2-72B: 29.1 (#310)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| WeirdML | 1.7% | 11.3% |
| BigCodeBench Instruct | 32.8% | 38.5% |
| LMArena Coding | 1195 | 1196 |
| BigCodeBench Complete | 40.5% | 54% |
| SciCode | 13.2% | — |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Llama 3.1-8B leads
Llama 3.1-8B: 22.5 (#131), Qwen2-72B: 17.0 (#146)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | — |
| TheAgentCompany | — | 1.1% |
| BALROG | 15.1% | — |
| METR Time Horizons | — | 29.9% |
Reasoning Qwen2-72B leads
Llama 3.1-8B: 14.9 (#321), Qwen2-72B: 23.2 (#181)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1175 | 1191 |
| Epoch Capabilities Index | 116.57 | 125.28 |
| CritPt | 0% | — |
| Chess Puzzles | 0% | — |
| DTBench | 50.9% | — |
| LMCA | 5.4% | — |
| PIQA | 81.2% | — |
Math Qwen2-72B leads
Llama 3.1-8B: 10.2 (#317), Qwen2-72B: 30.2 (#236)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Math | 1179 | 1235 |
| MATH Level 5 | 22.9% | 39.1% |
| OTIS Mock AIME 2024-2025 | 1.7% | — |
| Omni-MATH | 13.7% | — |
| GSM8K | 82.4% | — |
Knowledge Qwen2-72B leads
Llama 3.1-8B: 8.0 (#307), Qwen2-72B: 21.2 (#275)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 27% | 40.8% |
| LMArena Expert | 1144 | 1171 |
| MMLU | 56.1% | 82.4% |
| MMLU-Pro | 40.6% | — |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
Multilingual Qwen2-72B leads
Llama 3.1-8B: 34.0 (#249), Qwen2-72B: 35.9 (#244)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1148 | 1176 |
| LMArena Chinese | 1151 | 1240 |
| LMArena French | 1177 | 1170 |
| LMArena German | 1144 | 1151 |
| LMArena Japanese | 1061 | 1111 |
| LMArena Korean | 1053 | 1083 |
| LMArena Russian | 1158 | 1169 |
| LMArena Spanish | 1169 | 1169 |
Instruction Following Qwen2-72B leads
Llama 3.1-8B: 58.9 (#258), Qwen2-72B: 61.7 (#241)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1159 | 1181 |
| IFEval | 74.3% | — |
Long Context Too close to call
Llama 3.1-8B: 35.8 (#238), Qwen2-72B: 36.1 (#235)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1182 | 1192 |
Writing & Preference Qwen2-72B leads
Llama 3.1-8B: 29.7 (#290), Qwen2-72B: 40.8 (#241)
| Benchmark | Llama 3.1-8B | Qwen2-72B |
|---|---|---|
| LMArena Text | 1187 | 1203 |
| LMArena Creative Writing | 1154 | 1181 |
| LMArena Multi-Turn | 1172 | 1196 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than Qwen2-72B?
Qwen2-72B is the stronger model overall, scoring 30.0 to 23.0 on the Noometry Index.
Is Llama 3.1-8B or Qwen2-72B better for coding?
Qwen2-72B scores higher on coding benchmarks: 29.1 versus 20.2 in the Noometry coding category.
How many benchmarks do Llama 3.1-8B and Qwen2-72B share?
24 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Qwen2-72B has 26.