Model comparison
Llama 3-8B vs Qwen2.5 72B Instruct
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 25.5 on the Noometry Index.
Last verified . 29 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 29 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and Qwen2.5 72B Instruct in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5 72B Instruct leads 27.0 to 7.8.
- The biggest single-benchmark swing is MATH Level 5: 6.1% for Llama 3-8B and 63.2% for Qwen2.5 72B Instruct.
Side by side
| Llama 3-8B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.5 | 31.9 |
| Released | 2024-04-18 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 34 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Llama 3-8B: 31.0 (#289), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 31.9% | 45.8% |
| LMArena Coding | 1152 | 1292 |
| BigCodeBench Complete | 36.9% | 55.9% |
| WeirdML | — | 16% |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3-8B: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Llama 3-8B: 14.3 (#326), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1133 | 1271 |
| DTBench | 43.9% | 62.9% |
| Epoch Capabilities Index | 116.45 | 129 |
| ForecastBench | 58.6 | 57.5 |
| WinoGrande | 75.7% | 82.3% |
| Chess Puzzles | 0% | — |
| LMCA | — | 13.4% |
| Adversarial NLI | 57.3% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
Math Qwen2.5 72B Instruct leads
Llama 3-8B: 8.8 (#323), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 8.1% |
| LMArena Math | 1151 | 1283 |
| MATH Level 5 | 6.1% | 63.2% |
| Omni-MATH | — | 33% |
Knowledge Qwen2.5 72B Instruct leads
Llama 3-8B: 7.8 (#308), Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 26.1% | 49.1% |
| LMArena Expert | 1113 | 1245 |
| ARC (AI2) Challenge | 82.8% | 94.5% |
| MMLU | 68.8% | 85.3% |
| TriviaQA | 67.7% | 71.9% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| OpenBookQA | 82.6% | — |
Multilingual Qwen2.5 72B Instruct leads
Llama 3-8B: 30.8 (#261), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1098 | 1252 |
| LMArena Chinese | 1076 | 1272 |
| LMArena French | 1159 | 1280 |
| LMArena German | 1104 | 1234 |
| LMArena Japanese | 967 | 1180 |
| LMArena Korean | 1004 | 1188 |
| LMArena Russian | 1109 | 1264 |
| LMArena Spanish | 1173 | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Llama 3-8B: 58.4 (#260), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1127 | 1254 |
| IFEval | — | 80.6% |
Long Context Qwen2.5 72B Instruct leads
Llama 3-8B: 34.2 (#251), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1128 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Llama 3-8B: 37.5 (#256), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Llama 3-8B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1166 | 1269 |
| LMArena Creative Writing | 1150 | 1221 |
| LMArena Multi-Turn | 1152 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Llama 3-8B better than Qwen2.5 72B Instruct?
Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 25.5 on the Noometry Index.
Is Llama 3-8B or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 31.0 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Qwen2.5 72B Instruct share?
29 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen2.5 72B Instruct has 43.