Model comparison
Hy3 vs Qwen3 14B
Hy3 is the stronger model overall, scoring 44.2 to 35.5 on the Noometry Index.
Last verified . 0 shared benchmarks.
Tencent
Alibaba (Qwen)Summary
- The widest gap is in coding, where Hy3 leads 46.8 to 37.3.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
- Hy3 accepts more context: 262K tokens versus 131K.
Side by side
| Hy3 | Qwen3 14B | |
|---|---|---|
| Provider | Tencent | Alibaba (Qwen) |
| Noometry Index | 44.2 | 35.5 |
| Released | 2026-07-06 | 2025-04 |
| Weights | Open | Open |
| Context window | 262K | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.0825 | $0.35 |
| Output $ / M tokens | $0.33 | $1.40 |
| Results tracked | 19 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Qwen3 14B: 37.3 (#195)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| LMArena WebDev | 1508 | — |
| SciCode | — | 31.6% |
| LMArena Coding | 1464 | — |
Agentic & Tool Use Not comparable
Hy3: —, Qwen3 14B: 29.6 (#83)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41% |
Reasoning Hy3 leads
Hy3: 26.1 (#136), Qwen3 14B: 18.5 (#280)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | — | 49.1% |
| NYT Connections (extended) | 41.2% | — |
| CritPt | — | 0% |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1447 | — |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
| Epoch Capabilities Index | — | 138.23 |
Math Hy3 leads
Hy3: 40.1 (#93), Qwen3 14B: 38.6 (#133)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.4% |
| LMArena Math | 1475 | — |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Qwen3 14B: 39.3 (#134)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| GPQA Diamond | — | 63.8% |
| Vectara Hallucination Rate | — | 5.4% |
| LMArena Expert | 1460 | — |
Multilingual Not comparable
Hy3: 53.5 (#65), Qwen3 14B: —
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1426 | — |
| LMArena Chinese | 1493 | — |
| LMArena French | 1461 | — |
| LMArena German | 1439 | — |
| LMArena Japanese | 1392 | — |
| LMArena Korean | 1395 | — |
| LMArena Russian | 1432 | — |
| LMArena Spanish | 1456 | — |
Instruction Following Not comparable
Hy3: 75.1 (#70), Qwen3 14B: —
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| LMArena Instruction Following | 1426 | — |
Long Context Hy3 leads
Hy3: 44.1 (#75), Qwen3 14B: 38.1 (#204)
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
| LMArena Longer Query | 1442 | — |
Writing & Preference Not comparable
Hy3: 62.2 (#81), Qwen3 14B: —
| Benchmark | Hy3 | Qwen3 14B |
|---|---|---|
| LMArena Text | 1439 | — |
| LMArena Creative Writing | 1402 | — |
| LMArena Multi-Turn | 1436 | — |
Frequently asked questions
Is Hy3 better than Qwen3 14B?
Hy3 is the stronger model overall, scoring 44.2 to 35.5 on the Noometry Index.
Which is cheaper, Hy3 or Qwen3 14B?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.
Is Hy3 or Qwen3 14B better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 37.3 in the Noometry coding category.
Which has the bigger context window?
Hy3 does, with 262K tokens against 131K.
How many benchmarks do Hy3 and Qwen3 14B share?
0 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Qwen3 14B has 12.