Model comparison
Llama 2-13B vs Qwen2.5-Coder-32B
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.6 on the Noometry Index.
Last verified . 18 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 18 benchmarks with published results for both. Llama 2-13B scores higher in 1 category and Qwen2.5-Coder-32B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5-Coder-32B leads 41.6 to 29.8.
Side by side
| Llama 2-13B | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 29.6 | 33.4 |
| Released | 2023-07-18 | 2024-09-18 |
| Weights | Open | Open |
| Context window | — | 33K |
| Max output | — | 29K |
| Input $ / M tokens | — | $0.66 |
| Output $ / M tokens | — | $1 |
| Results tracked | 32 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-13B leads
Llama 2-13B: 30.9 (#291), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1062 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Qwen2.5-Coder-32B leads
Llama 2-13B: 12.8 (#337), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1051 | 1251 |
| Epoch Capabilities Index | 106.17 | 119.49 |
| HellaSwag | 80.7% | 83% |
| WinoGrande | 72.8% | 80.8% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 42.1% |
| DTBench | 42.2% | — |
| LiveBench Data Analysis | — | 49.9% |
| BIG-Bench Hard | 58.2% | — |
| LAMBADA | 76.5% | — |
| LiveBench | — | 46.2% |
| PIQA | 80.8% | — |
Math Qwen2.5-Coder-32B leads
Llama 2-13B: 31.1 (#229), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1065 | 1251 |
| GSM8K | 36.9% | 93% |
| LiveBench Math | — | 46.6% |
Knowledge Qwen2.5-Coder-32B leads
Llama 2-13B: 28.1 (#249), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1030 | 1221 |
| ARC (AI2) Challenge | 60.3% | 70.5% |
| MMLU | 55.6% | 79.1% |
| BoolQ | 82.4% | — |
| OpenBookQA | 57% | — |
| TriviaQA | 79.6% | — |
Multimodal Not comparable
Llama 2-13B: —, Qwen2.5-Coder-32B: —
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| ScienceQA | 55.8% | — |
Multilingual Qwen2.5-Coder-32B leads
Llama 2-13B: 26.5 (#279), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1024 | 1205 |
| LMArena Chinese | 1001 | 1222 |
| LMArena Russian | 1055 | 1228 |
| LMArena French | 1044 | — |
| LMArena German | 1009 | — |
| LMArena Japanese | 894 | — |
| LMArena Korean | 953 | — |
| LMArena Spanish | 1087 | — |
Instruction Following Qwen2.5-Coder-32B leads
Llama 2-13B: 53.3 (#287), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1045 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Qwen2.5-Coder-32B leads
Llama 2-13B: 32.3 (#269), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1064 | 1251 |
Writing & Preference Qwen2.5-Coder-32B leads
Llama 2-13B: 29.8 (#289), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Llama 2-13B | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1084 | 1230 |
| LMArena Creative Writing | 1047 | 1174 |
| LMArena Multi-Turn | 1050 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Llama 2-13B better than Qwen2.5-Coder-32B?
Qwen2.5-Coder-32B is the stronger model overall, scoring 33.4 to 29.6 on the Noometry Index.
Is Llama 2-13B or Qwen2.5-Coder-32B better for coding?
Llama 2-13B scores higher on coding benchmarks: 30.9 versus 22.6 in the Noometry coding category.
How many benchmarks do Llama 2-13B and Qwen2.5-Coder-32B share?
18 benchmarks have published results for both models. Llama 2-13B has 32 scored results on Noometry and Qwen2.5-Coder-32B has 31.