Model comparison
Codellama 34b Instruct vs Qwen1.5-14B
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 30.8 on the Noometry Index.
Last verified . 10 shared benchmarks.
Meta
Alibaba (Qwen)Summary
- They share 10 benchmarks with published results for both. Codellama 34b Instruct scores higher in 0 categories and Qwen1.5-14B in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen1.5-14B leads 33.6 to 28.2.
Side by side
| Codellama 34b Instruct | Qwen1.5-14B | |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 30.8 | 32.7 |
| Released | — | 2024-02-04 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 14 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen1.5-14B leads
Codellama 34b Instruct: 28.5 (#314), Qwen1.5-14B: 33.1 (#263)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Coding | 1046 | 1138 |
| BigCodeBench Instruct | 29% | — |
| BigCodeBench Complete | 37.1% | — |
| HumanEval+ | 43.9% | — |
| MBPP+ | 56.3% | — |
Reasoning Qwen1.5-14B leads
Codellama 34b Instruct: 19.6 (#255), Qwen1.5-14B: 21.4 (#223)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Hard Prompts | 1032 | 1113 |
Math Qwen1.5-14B leads
Codellama 34b Instruct: 31.0 (#230), Qwen1.5-14B: 32.4 (#215)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Math | 1056 | 1125 |
Knowledge Not comparable
Codellama 34b Instruct: —, Qwen1.5-14B: 29.8 (#232)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Expert | — | 1094 |
| MMLU | — | 68.6% |
Multilingual Qwen1.5-14B leads
Codellama 34b Instruct: 25.8 (#284), Qwen1.5-14B: 30.7 (#262)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Non-English | 1011 | 1095 |
| LMArena Chinese | 976 | 1147 |
| LMArena French | — | 1116 |
| LMArena German | — | 1043 |
| LMArena Japanese | — | 1019 |
| LMArena Russian | — | 1046 |
| LMArena Spanish | — | 1085 |
Instruction Following Qwen1.5-14B leads
Codellama 34b Instruct: 52.2 (#291), Qwen1.5-14B: 56.8 (#271)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Instruction Following | 1028 | 1102 |
Long Context Qwen1.5-14B leads
Codellama 34b Instruct: 30.9 (#284), Qwen1.5-14B: 33.7 (#257)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Longer Query | 1013 | 1113 |
Writing & Preference Qwen1.5-14B leads
Codellama 34b Instruct: 28.2 (#297), Qwen1.5-14B: 33.6 (#276)
| Benchmark | Codellama 34b Instruct | Qwen1.5-14B |
|---|---|---|
| LMArena Text | 1066 | 1128 |
| LMArena Creative Writing | 1032 | 1091 |
| LMArena Multi-Turn | 1015 | 1110 |
Frequently asked questions
Is Codellama 34b Instruct better than Qwen1.5-14B?
Qwen1.5-14B is the stronger model overall, scoring 32.7 to 30.8 on the Noometry Index.
Is Codellama 34b Instruct or Qwen1.5-14B better for coding?
Qwen1.5-14B scores higher on coding benchmarks: 33.1 versus 28.5 in the Noometry coding category.
How many benchmarks do Codellama 34b Instruct and Qwen1.5-14B share?
10 benchmarks have published results for both models. Codellama 34b Instruct has 14 scored results on Noometry and Qwen1.5-14B has 17.