Model comparison
Hy3 vs Step 3.5 Flash
Hy3 is the stronger model overall, scoring 44.2 to 42.3 on the Noometry Index.
Last verified . 18 shared benchmarks.
Tencent
StepFunSummary
- They share 18 benchmarks with published results for both. Hy3 scores higher in 7 categories and Step 3.5 Flash in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Hy3 leads 46.8 to 42.4.
- The biggest single-benchmark swing is NYT Connections (extended): 41.2% for Hy3 and 28.4% for Step 3.5 Flash.
- Both cost about the same: $0.0825 input and $0.33 output per million tokens.
- Hy3 accepts more context: 262K tokens versus 256K.
Side by side
| Hy3 | Step 3.5 Flash | |
|---|---|---|
| Provider | Tencent | StepFun |
| Noometry Index | 44.2 | 42.3 |
| Released | 2026-07-06 | 2026-01-29 |
| Weights | Open | Open |
| Context window | 262K | 256K |
| Max output | 128K | 256K |
| Input $ / M tokens | $0.0825 | $0.10 |
| Output $ / M tokens | $0.33 | $0.30 |
| Results tracked | 19 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Hy3 leads
Hy3: 46.8 (#63), Step 3.5 Flash: 42.4 (#105)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Coding | 1464 | 1436 |
| LMArena WebDev | 1508 | — |
Reasoning Hy3 leads
Hy3: 26.1 (#136), Step 3.5 Flash: 22.2 (#202)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| NYT Connections (extended) | 41.2% | 28.4% |
| LMArena Hard Prompts | 1447 | 1411 |
Math Step 3.5 Flash leads
Hy3: 40.1 (#93), Step 3.5 Flash: 42.6 (#84)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Math | 1475 | 1408 |
| MathArena Final-Answer Competitions | — | 66.8% |
Knowledge Hy3 leads
Hy3: 40.8 (#114), Step 3.5 Flash: 39.6 (#132)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Expert | 1460 | 1421 |
Multilingual Hy3 leads
Hy3: 53.5 (#65), Step 3.5 Flash: 50.5 (#119)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Non-English | 1426 | 1385 |
| LMArena Chinese | 1493 | 1447 |
| LMArena French | 1461 | 1421 |
| LMArena German | 1439 | 1405 |
| LMArena Japanese | 1392 | 1354 |
| LMArena Korean | 1395 | 1352 |
| LMArena Russian | 1432 | 1385 |
| LMArena Spanish | 1456 | 1419 |
Instruction Following Hy3 leads
Hy3: 75.1 (#70), Step 3.5 Flash: 73.1 (#124)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Instruction Following | 1426 | 1385 |
Long Context Hy3 leads
Hy3: 44.1 (#75), Step 3.5 Flash: 42.8 (#117)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Longer Query | 1442 | 1402 |
Writing & Preference Hy3 leads
Hy3: 62.2 (#81), Step 3.5 Flash: 58.8 (#113)
| Benchmark | Hy3 | Step 3.5 Flash |
|---|---|---|
| LMArena Text | 1439 | 1403 |
| LMArena Creative Writing | 1402 | 1357 |
| LMArena Multi-Turn | 1436 | 1405 |
Frequently asked questions
Is Hy3 better than Step 3.5 Flash?
Hy3 is the stronger model overall, scoring 44.2 to 42.3 on the Noometry Index.
Which is cheaper, Hy3 or Step 3.5 Flash?
Hy3 is cheaper. It lists at $0.0825 per million input tokens and $0.33 per million output tokens; Step 3.5 Flash lists at $0.10 and $0.30.
Is Hy3 or Step 3.5 Flash better for coding?
Hy3 scores higher on coding benchmarks: 46.8 versus 42.4 in the Noometry coding category.
Which has the bigger context window?
Hy3 does, with 262K tokens against 256K.
How many benchmarks do Hy3 and Step 3.5 Flash share?
18 benchmarks have published results for both models. Hy3 has 19 scored results on Noometry and Step 3.5 Flash has 19.