Model comparison
Hy3 vs Qwen2.5-Coder-32B
Hy3 is the stronger model overall, scoring 44.2 to 33.4 on the Noometry Index.
Last verified . 12 shared benchmarks.
Tencent
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. Hy3 scores higher in 8 categories and Qwen2.5-Coder-32B in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Hy3 leads 46.8 to 22.6.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- Hy3 accepts more context: 262K tokens versus 33K.
Side by side
| Hy3 | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | Tencent | Alibaba (Qwen) |
| Noometry Index | 44.2 | 33.4 |
| Released | 2026-07-06 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 262K | 33K |
| Max output | 128K | 29K |
| Input $ / M tokens | $0.0825 | $0.66 |
| Output $ / M tokens | $0.33 | $1 |
| Results tracked | 19 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Coding | 1464 | 1276 |
| SWE-bench Verified (bash only) | — | 9% |
| Aider Polyglot | — | 16.4% |
| LMArena WebDev | 1508 | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Reasoning Hy3 leads
Hy3: 26.1 (#136), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1447 | 1251 |
| NYT Connections (extended) | 41.2% | — |
| LiveBench Reasoning | — | 42.1% |
| LiveBench Data Analysis | — | 49.9% |
| Epoch Capabilities Index | — | 119.49 |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math Hy3 leads
Hy3: 40.1 (#93), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1475 | 1251 |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1460 | 1221 |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual Hy3 leads
Hy3: 53.5 (#65), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1426 | 1205 |
| LMArena Chinese | 1493 | 1222 |
| LMArena Russian | 1432 | 1228 |
| LMArena French | 1461 | — |
| LMArena German | 1439 | — |
| LMArena Japanese | 1392 | — |
| LMArena Korean | 1395 | — |
| LMArena Spanish | 1456 | — |
Instruction Following Hy3 leads
Hy3: 75.1 (#70), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1426 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
Long Context Hy3 leads
Hy3: 44.1 (#75), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1442 | 1251 |
Writing & Preference Hy3 leads
Hy3: 62.2 (#81), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | Hy3 | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1439 | 1230 |
| LMArena Creative Writing | 1402 | 1174 |
| LMArena Multi-Turn | 1436 | 1222 |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is Hy3 better than Qwen2.5-Coder-32B?
Hy3 is the stronger model overall, scoring 44.2 to 33.4 on the Noometry Index.
Which is cheaper, Hy3 or Qwen2.5-Coder-32B?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is Hy3 or Qwen2.5-Coder-32B better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
Hy3 does, with 262K tokens against 33K.
How many benchmarks do Hy3 and Qwen2.5-Coder-32B share?
12 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Qwen2.5-Coder-32B has 31.