Model comparison
Llama 3-8B vs Qwen1.5-110B
Qwen1.5-110B is the stronger model overall, scoring 33.6 to 25.3 on the Noometry Index.
Last verified . 20 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. Llama 3-8B scores higher in 0 categories and Qwen1.5-110B in 8 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen1.5-110B leads 30.7 to 7.8.
- The biggest single-benchmark swing is BigCodeBench Complete: 36.9% for Llama 3-8B and 44.4% for Qwen1.5-110B.
Side by side
| Llama 3-8B | Qwen1.5-110B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 25.3 | 33.6 |
| Released | 2024-04-18 | 2024-04-25 |
| Weights | Open | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.14 | — |
| Output $ / M tokens | $0.14 | — |
| Results tracked | 34 | 20 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-110B leads
Llama 3-8B: 30.5 (#285), Qwen1.5-110B: 32.4 (#260)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| BigCodeBench Instruct | 31.9% | 35% |
| LMArena Coding | 1152 | 1184 |
| BigCodeBench Complete | 36.9% | 44.4% |
| HumanEval+ | 56.7% | — |
| MBPP+ | 54.8% | — |
Reasoning Qwen1.5-110B leads
Llama 3-8B: 13.8 (#321), Qwen1.5-110B: 21.7 (#192)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Hard Prompts | 1133 | 1168 |
| ForecastBench | 58.6 | 57.7 |
| Chess Puzzles | 0% | — |
| DTBench | 43.9% | — |
| Adversarial NLI | 57.3% | — |
| Epoch Capabilities Index | 116.33 | — |
| WinoGrande | 75.7% | — |
Math Qwen1.5-110B leads
Llama 3-8B: 8.3 (#317), Qwen1.5-110B: 31.1 (#199)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Math | 1151 | 1185 |
| OTIS Mock AIME 2024-2025 | 1.9% | — |
| MATH Level 5 | 6.1% | — |
Knowledge Qwen1.5-110B leads
Llama 3-8B: 7.8 (#303), Qwen1.5-110B: 30.7 (#217)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Expert | 1113 | 1144 |
| GPQA Diamond | 26.1% | — |
| ARC (AI2) Challenge | 82.8% | — |
| MMLU | 68.8% | — |
| OpenBookQA | 82.6% | — |
| TriviaQA | 67.7% | — |
Multilingual Qwen1.5-110B leads
Llama 3-8B: 31.0 (#256), Qwen1.5-110B: 33.8 (#245)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Non-English | 1098 | 1142 |
| LMArena Chinese | 1076 | 1206 |
| LMArena French | 1158 | 1150 |
| LMArena German | 1104 | 1123 |
| LMArena Japanese | 967 | 1074 |
| LMArena Korean | 1003 | 1044 |
| LMArena Russian | 1109 | 1118 |
| LMArena Spanish | 1172 | 1141 |
Instruction Following Qwen1.5-110B leads
Llama 3-8B: 58.7 (#255), Qwen1.5-110B: 60.6 (#247)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Instruction Following | 1127 | 1158 |
Long Context Too close to call
Llama 3-8B: 34.2 (#246), Qwen1.5-110B: 35.0 (#237)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Longer Query | 1128 | 1157 |
Writing & Preference Too close to call
Llama 3-8B: 37.7 (#251), Qwen1.5-110B: 38.1 (#250)
| Benchmark | Llama 3-8B | Qwen1.5-110B |
|---|---|---|
| LMArena Text | 1166 | 1175 |
| LMArena Creative Writing | 1150 | 1148 |
| LMArena Multi-Turn | 1152 | 1160 |
Frequently asked questions
Is Llama 3-8B better than Qwen1.5-110B?
Qwen1.5-110B is the stronger model overall, scoring 33.6 to 25.3 on the Noometry Index.
Is Llama 3-8B or Qwen1.5-110B better for coding?
Qwen1.5-110B scores higher on coding benchmarks: 32.4 versus 30.5 in the Noometry coding category.
How many benchmarks do Llama 3-8B and Qwen1.5-110B share?
20 benchmarks have published results for both models. Llama 3-8B has 34 scored results on Noometry and Qwen1.5-110B has 20.