Model comparison
Qwen-14B vs Qwen2.5 72B Instruct
Qwen-14B and Qwen2.5 72B Instruct score almost the same on the Noometry Index (31.4 vs 31.9), so choose on price, context window or the category you care about most.
Last verified . 15 shared benchmarks.
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. Qwen-14B scores higher in 1 category and Qwen2.5 72B Instruct in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen2.5 72B Instruct leads 46.7 to 27.6.
Side by side
| Qwen-14B | Qwen2.5 72B Instruct | |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.4 | 31.9 |
| Released | 2023-09-24 | 2024-09 |
| Weights | Open | Open |
| Context window | — | 131K |
| Max output | — | 8K |
| Input $ / M tokens | — | $1.40 |
| Output $ / M tokens | — | $5.60 |
| Results tracked | 18 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5 72B Instruct leads
Qwen-14B: 31.2 (#288), Qwen2.5 72B Instruct: 33.2 (#260)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1071 | 1292 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
Agentic & Tool Use Not comparable
Qwen-14B: —, Qwen2.5 72B Instruct: 22.1 (#133)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
Reasoning Qwen2.5 72B Instruct leads
Qwen-14B: 19.6 (#257), Qwen2.5 72B Instruct: 22.3 (#199)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1027 | 1271 |
| BIG-Bench Hard | 55% | 79.8% |
| Epoch Capabilities Index | 113.03 | 129 |
| PIQA | 79.9% | 82.6% |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| LAMBADA | 71.1% | — |
| WinoGrande | — | 82.3% |
Math Qwen-14B leads
Qwen-14B: 31.2 (#227), Qwen2.5 72B Instruct: 19.3 (#287)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1068 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |
| GSM8K | 61.3% | — |
Knowledge Not comparable
Qwen-14B: —, Qwen2.5 72B Instruct: 27.0 (#253)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| ARC (AI2) Challenge | 84.4% | 94.5% |
| MMLU | 66.3% | 85.3% |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| BoolQ | 86.2% | — |
| TriviaQA | — | 71.9% |
Multilingual Qwen2.5 72B Instruct leads
Qwen-14B: 27.5 (#275), Qwen2.5 72B Instruct: 41.0 (#213)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1041 | 1252 |
| LMArena Chinese | 1077 | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |
Instruction Following Qwen2.5 72B Instruct leads
Qwen-14B: 52.4 (#289), Qwen2.5 72B Instruct: 65.5 (#221)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1031 | 1254 |
| IFEval | — | 80.6% |
Long Context Qwen2.5 72B Instruct leads
Qwen-14B: 31.3 (#280), Qwen2.5 72B Instruct: 38.9 (#188)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1028 | 1282 |
Writing & Preference Qwen2.5 72B Instruct leads
Qwen-14B: 27.6 (#299), Qwen2.5 72B Instruct: 46.7 (#215)
| Benchmark | Qwen-14B | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1051 | 1269 |
| LMArena Creative Writing | 1028 | 1221 |
| LMArena Multi-Turn | 1022 | 1272 |
| WildBench | — | 80.2% |
Frequently asked questions
Is Qwen-14B better than Qwen2.5 72B Instruct?
Qwen-14B and Qwen2.5 72B Instruct score almost the same on the Noometry Index (31.4 vs 31.9), so choose on price, context window or the category you care about most.
Is Qwen-14B or Qwen2.5 72B Instruct better for coding?
Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 31.2 in the Noometry coding category.
How many benchmarks do Qwen-14B and Qwen2.5 72B Instruct share?
15 benchmarks have published results for both models. Qwen-14B has 18 scored results on Noometry and Qwen2.5 72B Instruct has 43.