Model comparison
Llama 13b vs Qwen1.5 4b Chat
Qwen1.5 4b Chat is the stronger model overall, scoring 28.3 to 23.9 on the Noometry Index.
Last verified . 8 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 8 benchmarks with published results for both. Llama 13b scores higher in 0 categories and Qwen1.5 4b Chat in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Qwen1.5 4b Chat leads 49.3 to 36.9.
Side by side
| Llama 13b | Qwen1.5 4b Chat | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 23.9 | 28.3 |
| Released | 2023-02-24 | — |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 21 | 13 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5 4b Chat leads
Llama 13b: 20.9 (#332), Qwen1.5 4b Chat: 28.6 (#305)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Coding | 683 | 999 |
Reasoning Qwen1.5 4b Chat leads
Llama 13b: 13.4 (#327), Qwen1.5 4b Chat: 17.7 (#277)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Hard Prompts | 728 | 976 |
| BIG-Bench Hard | 37.9% | — |
| Epoch Capabilities Index | 100.09 | — |
| HellaSwag | 79.2% | — |
| LAMBADA | 75.2% | — |
| PIQA | 80.1% | — |
| WinoGrande | 73% | — |
Math Qwen1.5 4b Chat leads
Llama 13b: 24.4 (#259), Qwen1.5 4b Chat: 27.9 (#233)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Math | 838 | 1026 |
| GSM8K | 20.6% | — |
Knowledge Not comparable
Llama 13b: —, Qwen1.5 4b Chat: 26.2 (#252)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Expert | — | 980 |
| ARC (AI2) Challenge | 52.7% | — |
| BoolQ | 78.7% | — |
| MMLU | 47.7% | — |
| OpenBookQA | 56.4% | — |
| TriviaQA | 77.9% | — |
Multimodal Not comparable
Llama 13b: —, Qwen1.5 4b Chat: —
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| ScienceQA | 43.3% | — |
Multilingual Qwen1.5 4b Chat leads
Llama 13b: 16.8 (#292), Qwen1.5 4b Chat: 24.2 (#285)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Non-English | 819 | 979 |
| LMArena Chinese | — | 1024 |
| LMArena German | — | 902 |
| LMArena Russian | — | 952 |
Instruction Following Qwen1.5 4b Chat leads
Llama 13b: 36.9 (#300), Qwen1.5 4b Chat: 49.3 (#295)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Instruction Following | 781 | 979 |
Long Context Not comparable
Llama 13b: —, Qwen1.5 4b Chat: 30.1 (#285)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Longer Query | — | 988 |
Writing & Preference Qwen1.5 4b Chat leads
Llama 13b: 13.9 (#307), Qwen1.5 4b Chat: 23.9 (#304)
| Benchmark | Llama 13b | Qwen1.5 4b Chat |
|---|---|---|
| LMArena Text | 834 | 997 |
| LMArena Creative Writing | 794 | 969 |
| LMArena Multi-Turn | 753 | 977 |
Frequently asked questions
Is Llama 13b better than Qwen1.5 4b Chat?
Qwen1.5 4b Chat is the stronger model overall, scoring 28.3 to 23.9 on the Noometry Index.
Is Llama 13b or Qwen1.5 4b Chat better for coding?
Qwen1.5 4b Chat scores higher on coding benchmarks: 28.6 versus 20.9 in the Noometry coding category.
How many benchmarks do Llama 13b and Qwen1.5 4b Chat share?
8 benchmarks have published results for both models. Llama 13b has 21 scored results on Noometry and Qwen1.5 4b Chat has 13.