Model comparison
Hy3 vs Hy4 preview
Hy4 preview is the stronger model overall, scoring 45.3 to 44.2 on the Noometry Index. Hy3 costs 7.8× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.
Last verified . 2 shared benchmarks.
Summary
- They share 2 benchmarks with published results for both. Hy3 scores higher in 0 categories and Hy4 preview in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Hy4 preview leads 55.7 to 40.1.
- The biggest single-benchmark swing is NYT Connections (extended): 41.2% for Hy3 and 68.2% for Hy4 preview.
- Hy3 is cheaper at $0.0825 / $0.33 per million input/output tokens, against $0.75 / $2.25 for Hy4 preview.
- Hy4 preview accepts more context: 1.05M tokens versus 262K.
Side by side
| Hy3 | Hy4 preview | |
|---|---|---|
| Provider | Tencent | Tencent |
| Noometry Index | 44.2 | 45.3 |
| Released | 2026-07-06 | 2026-08-28 |
| Weights | Open | Open |
| Context window | 262K | 1.05M |
| Max output | 128K | 64K |
| Input $ / M tokens | $0.0825 | $0.75 |
| Output $ / M tokens | $0.33 | $2.25 |
| Results tracked | 19 | 3 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy4 preview leads
Hy3: 46.8 (#63), Hy4 preview: 51.6 (#38)
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| LMArena WebDev | 1508 | 1632 |
| LMArena Coding | 1464 | — |
Reasoning Hy4 preview leads
Hy3: 26.1 (#136), Hy4 preview: 31.9 (#79)
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| NYT Connections (extended) | 41.2% | 68.2% |
| LMArena Hard Prompts | 1447 | — |
Math Hy4 preview leads
Hy3: 40.1 (#93), Hy4 preview: 55.7 (#42)
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| ProofBench | — | 75% |
| LMArena Math | 1475 | — |
Knowledge Not comparable
Hy3: 40.8 (#114), Hy4 preview: —
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| LMArena Expert | 1460 | — |
Multilingual Not comparable
Hy3: 53.5 (#65), Hy4 preview: —
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| 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), Hy4 preview: —
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| LMArena Instruction Following | 1426 | — |
Long Context Not comparable
Hy3: 44.1 (#75), Hy4 preview: —
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| LMArena Longer Query | 1442 | — |
Writing & Preference Not comparable
Hy3: 62.2 (#81), Hy4 preview: —
| Benchmark | Hy3 | Hy4 preview |
|---|---|---|
| LMArena Text | 1439 | — |
| LMArena Creative Writing | 1402 | — |
| LMArena Multi-Turn | 1436 | — |
Frequently asked questions
Is Hy3 better than Hy4 preview?
Hy4 preview is the stronger model overall, scoring 45.3 to 44.2 on the Noometry Index. Hy3 costs 7.8× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.
Which is cheaper, Hy3 or Hy4 preview?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Hy4 preview lists at $0.75 and $2.25.
Is Hy3 or Hy4 preview better for coding?
Hy4 preview scores higher on coding benchmarks: 51.6 versus 46.8 in the Noometry coding category.
Which has the bigger context window?
Hy4 preview does, with 1.05M tokens against 262K.
How many benchmarks do Hy3 and Hy4 preview share?
2 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Hy4 preview has 3.