Model comparison
Hy3 vs Qwen2.5-Max
Hy3 is the stronger model overall, scoring 44.2 to 40.7 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 Qwen2.5-Max in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Hy3 leads 62.2 to 55.4.
- Hy3 has downloadable open weights; the other is API-only.
Side by side
| Hy3 | Qwen2.5-Max | |
|---|---|---|
| Provider | Tencent | Alibaba (Qwen) |
| Noometry Index | 44.2 | 40.7 |
| Released | 2026-07-06 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.13 | — |
| Output $ / M tokens | $0.53 | — |
| Results tracked | 19 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Qwen2.5-Max: 41.8 (#117)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1464 | 1359 |
| LMArena WebDev | 1508 | — |
| LiveBench Coding | — | 64.4% |
Reasoning Too close to call
Hy3: 26.1 (#136), Qwen2.5-Max: 25.6 (#147)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1360 |
| NYT Connections (extended) | 41.2% | — |
| LiveBench Reasoning | — | 51.4% |
| LiveBench Data Analysis | — | 67.9% |
| Epoch Capabilities Index | — | 132.53 |
| LiveBench | — | 62.3% |
Math Hy3 leads
Hy3: 40.1 (#93), Qwen2.5-Max: 36.9 (#162)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1475 | 1369 |
| LiveBench Math | — | 58.4% |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Qwen2.5-Max: 35.3 (#186)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1460 | 1337 |
| Confabulations | — | 21.8% |
Multilingual Hy3 leads
Hy3: 53.5 (#65), Qwen2.5-Max: 48.1 (#146)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1426 | 1352 |
| LMArena Chinese | 1493 | 1382 |
| LMArena French | 1461 | 1396 |
| LMArena German | 1439 | 1350 |
| LMArena Japanese | 1392 | 1300 |
| LMArena Korean | 1395 | 1304 |
| LMArena Russian | 1432 | 1353 |
| LMArena Spanish | 1456 | 1377 |
Instruction Following Hy3 leads
Hy3: 75.1 (#70), Qwen2.5-Max: 71.3 (#152)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1426 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
Long Context Hy3 leads
Hy3: 44.1 (#75), Qwen2.5-Max: 41.4 (#142)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1442 | 1358 |
Writing & Preference Hy3 leads
Hy3: 62.2 (#81), Qwen2.5-Max: 55.4 (#146)
| Benchmark | Hy3 | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1439 | 1367 |
| LMArena Creative Writing | 1402 | 1339 |
| LMArena Multi-Turn | 1436 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is Hy3 better than Qwen2.5-Max?
Hy3 is the stronger model overall, scoring 44.2 to 40.7 on the Noometry Index.
Is Hy3 or Qwen2.5-Max better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 41.8 in the Noometry coding category.
How many benchmarks do Hy3 and Qwen2.5-Max share?
17 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Qwen2.5-Max has 27.