Model comparison
Llama 3.2 3B vs Qwen1.5-14B
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 28.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 13 benchmarks with published results for both. Llama 3.2 3B scores higher in 1 category and Qwen1.5-14B in 7 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen1.5-14B leads 33.6 to 24.7.
Side by side
| Llama 3.2 3B | Qwen1.5-14B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 28.9 | 32.7 |
| Released | 2024-09-24 | 2024-02-04 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 118K | — |
| Input $ / M tokens | $0.05 | — |
| Output $ / M tokens | $0.33 | — |
| Results tracked | 18 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-14B leads
Llama 3.2 3B: 27.6 (#319), Qwen1.5-14B: 33.1 (#263)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 1098 | 1138 |
| BigCodeBench Instruct | 23.4% | — |
| BigCodeBench Complete | 28.3% | — |
Agentic & Tool Use Not comparable
Llama 3.2 3B: 20.1 (#143), Qwen1.5-14B: —
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 21.9% | — |
| BALROG | 10.1% | — |
Reasoning Too close to call
Llama 3.2 3B: 21.0 (#228), Qwen1.5-14B: 21.4 (#223)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 1095 | 1113 |
Math Too close to call
Llama 3.2 3B: 32.4 (#214), Qwen1.5-14B: 32.4 (#215)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 1126 | 1125 |
Knowledge Too close to call
Llama 3.2 3B: 29.7 (#235), Qwen1.5-14B: 29.8 (#232)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Expert | 1090 | 1094 |
| MMLU | — | 68.6% |
Multilingual Qwen1.5-14B leads
Llama 3.2 3B: 26.2 (#281), Qwen1.5-14B: 30.7 (#262)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 1019 | 1095 |
| LMArena Chinese | 1017 | 1147 |
| LMArena German | 1056 | 1043 |
| LMArena Russian | 949 | 1046 |
| LMArena French | — | 1116 |
| LMArena Japanese | — | 1019 |
| LMArena Spanish | — | 1085 |
Instruction Following Too close to call
Llama 3.2 3B: 56.0 (#275), Qwen1.5-14B: 56.8 (#271)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 1089 | 1102 |
Long Context Too close to call
Llama 3.2 3B: 33.4 (#261), Qwen1.5-14B: 33.7 (#257)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | 1100 | 1113 |
Writing & Preference Qwen1.5-14B leads
Llama 3.2 3B: 24.7 (#307), Qwen1.5-14B: 33.6 (#276)
| Benchmark | Llama 3.2 3B | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 1110 | 1128 |
| LMArena Creative Writing | 1094 | 1091 |
| LMArena Multi-Turn | 1105 | 1110 |
| EQ-Bench Creative Writing | 595 | — |
Frequently asked questions
Is Llama 3.2 3B better than Qwen1.5-14B?
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 28.9 on the Noometry Index.
Is Llama 3.2 3B or Qwen1.5-14B better for coding?
Qwen1.5-14B scores higher on coding benchmarks: 33.1 versus 27.6 in the Noometry coding category.
How many benchmarks do Llama 3.2 3B and Qwen1.5-14B share?
13 benchmarks have published results for both models. Llama 3.2 3B has 18 scored results on Noometry and Qwen1.5-14B has 17.