Hy3 vs Hy4 preview
Updated October 21, 2026. 3 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Hy4 preview for coding: it scores 61.6 on the Vibe Coding Index against 43.9 for Hy3.
Pick Hy3 when output tokens cost 4.7x less.
- Shared results
- 3
- Hy3 only
- 41
- Hy4 preview only
- 21
- Like-for-like categories
- 1 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, four public coding benchmarks
Hy4 preview
Hy4 preview scores 61.6 against 43.9.
One benchmark behind each score
Web design
Design Arena: websites
No call
Only Hy3 has a Design Arena: websites result, 1230.
Directional only
General
Vals Index
No call
Only Hy4 preview has a Vals Index result, 53.2%.
Directional only
Long documents
Listed context window
Hy4 preview
Hy4 preview takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
Hy3 $0.0004, Hy4 preview $0.0021 per turn.
List prices
At a glance
- Coding index
- Hy3 43.9 from 1 benchmark, Hy4 preview 61.6 from 1 benchmark.
- Widest category
- Coding, 17.1 points apart on average over 3 shared benchmarks.
- Widest single gap
- LMArena WebDev: HTML: Hy3 1487 against Hy4 preview 1620, 18.9 points apart on a 0 to 100 scale.
- Output price
- Hy3 $0.528 against Hy4 preview $2.501 per million output tokens, 4.7x apart.
- Context window
- Hy3 takes 262K tokens, Hy4 preview 1.05M.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Hy3 | Bars | Hy4 preview | Ties |
|---|---|---|---|---|
| Coding | 0 | 3 | ||
| All shared boards | 0 | 3 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index03
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| LMArena WebDev | 1507 | 1631 | Hy4 preview |
| Terminal-Bench 4 (Vals AI) | Not measured | 8.1% | Only Hy4 preview measured |
| Vibe Code Bench (Vals AI) | Not measured | 77.5% | Only Hy4 preview measured |
| Code migration (Vals AI) | Not measured | 47.4% | Only Hy4 preview measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only Hy4 preview measured |
| IOI (Vals AI) | Not measured | 59.3% | Only Hy4 preview measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 55.1% | Only Hy4 preview measured |
| CyberBench (Vals AI) | Not measured | 65.4% | Only Hy4 preview measured |
| SciCode | 48.6% | Not measured | Only Hy3 measured |
| ALE-Bench | 838.6 | Not measured | Only Hy3 measured |
| LMArena Text: coding | 1466 | Not measured | Only Hy3 measured |
| LMArena Text: software and IT | 1463 | Not measured | Only Hy3 measured |
| LMArena WebDev: React | 1511 | 1653 | Hy4 preview |
| LMArena WebDev: HTML | 1487 | 1620 | Hy4 preview |
| Neither model has a result on the other 24 boards here | |||
IntelligenceNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| Neither model has a result on 2 boards here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| SREBench (Vals AI) | Not measured | 2.3% | Only Hy4 preview measured |
| Finance Agent (Vals AI) | Not measured | 55.1% | Only Hy4 preview measured |
| Tax Agent Bench (Vals AI) | Not measured | 32.6% | Only Hy4 preview measured |
| Legal Research Bench (Vals AI) | Not measured | 45.2% | Only Hy4 preview measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 8.8% | Only Hy4 preview measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 58.1% | Only Hy4 preview measured |
| BioMysteryBench (Vals AI) | Not measured | 69.3% | Only Hy4 preview measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 1.4% | Only Hy4 preview measured |
| Neither model has a result on the other 31 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 89.7% | Not measured | Only Hy3 measured |
| Humanity's Last Exam (Artificial Analysis run) | 33.5% | Not measured | Only Hy3 measured |
| LMArena Text: hard prompts | 1447 | Not measured | Only Hy3 measured |
| LMArena Text: expert | 1461 | Not measured | Only Hy3 measured |
| LMArena Text: hard prompts in English | 1455 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 16 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| ProofBench | Not measured | 75.0% | Only Hy4 preview measured |
| LMArena Text: math | 1473 | Not measured | Only Hy3 measured |
| LMArena Text: mathematical work | 1482 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 10 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | Not measured | 83.8% | Only Hy4 preview measured |
| MedCode (Vals AI) | Not measured | 43.3% | Only Hy4 preview measured |
| LMArena Text: science | 1457 | Not measured | Only Hy3 measured |
| LMArena Text: medicine and healthcare | 1452 | Not measured | Only Hy3 measured |
| LMArena Text: legal and government | 1422 | Not measured | Only Hy3 measured |
| LMArena Text: business and finance | 1424 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 11 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| Design Arena: websites | 1230 | Not measured | Only Hy3 measured |
| Design Arena: UI components | 1205 | Not measured | Only Hy3 measured |
| Design Arena: game dev | 1208 | Not measured | Only Hy3 measured |
| Design Arena: data viz | 1163 | Not measured | Only Hy3 measured |
| Design Arena: 3D | 1233 | Not measured | Only Hy3 measured |
| Design Arena: all code categories | 1227 | Not measured | Only Hy3 measured |
| Design Arena: SVG | 1199 | Not measured | Only Hy3 measured |
| Design Arena: ASCII art | 1157 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 1 board here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index21 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| Vals Index | Not measured | 53.2% | Only Hy4 preview measured |
| Public Benefits Bench (Vals AI) | Not measured | 68.6% | Only Hy4 preview measured |
| MedScribe (Vals AI) | Not measured | 83.6% | Only Hy4 preview measured |
| Long context reasoning (AA-LCR) | 79.0% | Not measured | Only Hy3 measured |
| LMArena Text | 1438 | Not measured | Only Hy3 measured |
| LMArena Text: instruction following | 1426 | Not measured | Only Hy3 measured |
| LMArena Text: creative writing | 1403 | Not measured | Only Hy3 measured |
| LMArena Text: English | 1448 | Not measured | Only Hy3 measured |
| LMArena Text: non-English | 1426 | Not measured | Only Hy3 measured |
| LMArena Text: Chinese | 1485 | Not measured | Only Hy3 measured |
| LMArena Text: German | 1449 | Not measured | Only Hy3 measured |
| LMArena Text: Japanese | 1397 | Not measured | Only Hy3 measured |
| LMArena Text: Korean | 1396 | Not measured | Only Hy3 measured |
| LMArena Text: French | 1456 | Not measured | Only Hy3 measured |
| LMArena Text: Spanish | 1457 | Not measured | Only Hy3 measured |
| LMArena Text: Russian | 1432 | Not measured | Only Hy3 measured |
| LMArena Text: Polish | 1447 | Not measured | Only Hy3 measured |
| LMArena Text: multi-turn | 1436 | Not measured | Only Hy3 measured |
| LMArena Text: longer queries | 1442 | Not measured | Only Hy3 measured |
| LMArena Text: writing and language | 1414 | Not measured | Only Hy3 measured |
| LMArena Text: entertainment and media | 1400 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 25 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Hy3 | Hy4 preview | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Hy3 costs less
Repository review
50K input + 3K output tokens
Hy3 costs less
Agent loop
200K input + 10K output tokens
Hy3 costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Tencent Hy3 | Tencent Hy4 preview |
|---|---|---|
| Maker | Tencent | Tencent |
| Weights | Open weights | Open weights |
| Released | July 6, 2026 | August 28, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.132 | $0.834 |
| Output $/Mtok | $0.528 | $2.501 |
| Coding index | 43.9 | 61.6 |
| Independent measurements behind it | 1 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Hy3 better than Hy4 preview for coding?
Pick Hy4 preview for coding: it scores 61.6 on the Vibe Coding Index against 43.9 for Hy3. Pick Hy3 when output tokens cost 4.7x less.
Which is cheaper, Hy3 or Hy4 preview?
Hy3 costs $0.132 per million input tokens and $0.528 per million output tokens; Hy4 preview costs $0.834 and $2.501. Cost per solved task is known for neither of them.
Which has the bigger context window, Hy3 or Hy4 preview?
Hy4 preview, with 1.05M tokens against 262K.
How many benchmarks is the Hy3 vs Hy4 preview comparison based on?
3 benchmarks have a result for both models as of October 21, 2026. Hy3's coding index rests on 1 independent benchmark and Hy4 preview's on 1. Where only one model has a result, the page says so instead of naming a winner.