Model comparison
Hy3 vs Qwen2-72B
Hy3 is the stronger model overall, scoring 44.2 to 30.0 on the Noometry Index.
Last verified . 17 shared benchmarks.
Tencent
Alibaba (Qwen)Summary
- They share 17 benchmarks with published results for both. Hy3 scores higher in 8 categories and Qwen2-72B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.2 to 40.8.
Side by side
| Hy3 | Qwen2-72B | |
|---|---|---|
| Provider | Tencent | Alibaba (Qwen) |
| Noometry Index | 44.2 | 30.0 |
| Released | 2026-07-06 | 2024-06-07 |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.0825 | — |
| Output $ / M tokens | $0.33 | — |
| Results tracked | 19 | 26 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Qwen2-72B: 29.1 (#310)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1464 | 1196 |
| LMArena WebDev | 1508 | — |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |
Agentic & Tool Use Not comparable
Hy3: —, Qwen2-72B: 17.0 (#146)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |
Reasoning Hy3 leads
Hy3: 26.1 (#136), Qwen2-72B: 23.2 (#181)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1191 |
| NYT Connections (extended) | 41.2% | — |
| Epoch Capabilities Index | — | 125.28 |
Math Hy3 leads
Hy3: 40.1 (#93), Qwen2-72B: 30.2 (#236)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Math | 1475 | 1235 |
| MATH Level 5 | — | 39.1% |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Qwen2-72B: 21.2 (#275)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Expert | 1460 | 1171 |
| GPQA Diamond | — | 40.8% |
| MMLU | — | 82.4% |
Multilingual Hy3 leads
Hy3: 53.5 (#65), Qwen2-72B: 35.9 (#244)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1426 | 1176 |
| LMArena Chinese | 1493 | 1240 |
| LMArena French | 1461 | 1170 |
| LMArena German | 1439 | 1151 |
| LMArena Japanese | 1392 | 1111 |
| LMArena Korean | 1395 | 1083 |
| LMArena Russian | 1432 | 1169 |
| LMArena Spanish | 1456 | 1169 |
Instruction Following Hy3 leads
Hy3: 75.1 (#70), Qwen2-72B: 61.7 (#241)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1181 |
Long Context Hy3 leads
Hy3: 44.1 (#75), Qwen2-72B: 36.1 (#235)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1442 | 1192 |
Writing & Preference Hy3 leads
Hy3: 62.2 (#81), Qwen2-72B: 40.8 (#241)
| Benchmark | Hy3 | Qwen2-72B |
|---|---|---|
| LMArena Text | 1439 | 1203 |
| LMArena Creative Writing | 1402 | 1181 |
| LMArena Multi-Turn | 1436 | 1196 |
Frequently asked questions
Is Hy3 better than Qwen2-72B?
Hy3 is the stronger model overall, scoring 44.2 to 30.0 on the Noometry Index.
Is Hy3 or Qwen2-72B better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 29.1 in the Noometry coding category.
How many benchmarks do Hy3 and Qwen2-72B share?
17 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Qwen2-72B has 26.