Hy3 vs Hy4
Hy3 costs 2.4x less per output token than Hy4 ($0.528 vs $1.2505 per million). Updated November 14, 2026; 7 benchmarks have a result for both.
Hy3 or Hy4: which to pick
Only Hy4 has a coding score, 65.2 on the Vibe Coding Index; Hy3 has none, so the two cannot be ranked on coding.
Pick Hy3 for output tokens 2.4x cheaper ($0.528 vs $1.2505 per million).
- Shared results
- 7
- Hy3 only
- 41
- Hy4 only
- 12
- Like-for-like categories
- 3 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
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, six public coding benchmarks
No call
Only Hy4 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
No call
Only Hy4 has a AA Intelligence Index result, 34.6.
Directional only
Web design
Design Arena: websites
Hy4
Hy4 leads, 1251 against 1221.
Same board, both measured
General
Vals Index
No call
Only Hy4 has a Vals Index result, 59.4%.
Directional only
Long documents
Listed context window
Hy4
Hy4 takes 1.05M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
Hy3 $0.0004, Hy4 $0.0010 per turn.
List prices
At a glance
- Coding index
- Hy3 not benchmarked, Hy4 65.2 from 2 benchmarks.
- Widest category
- Coding, 36.9 points apart on average over 1 shared benchmark.
- Widest single gap
- Terminal-Bench 4.0 (Artificial Analysis run): Hy3 0.5% against Hy4 37.4%, 36.9 points apart on a 0 to 100 scale.
- Output price
- Hy3 $0.528 against Hy4 $1.2505 per million output tokens, 2.4x apart.
- Context window
- Hy3 takes 262K tokens, Hy4 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 | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| Reasoning | 0 | 1 | ||
| Web design | 0 | 5 | ||
| All shared boards | 0 | 7 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| LMArena WebDev | 1505 | Not measured | Only Hy3 measured |
| Terminal-Bench 4 (Vals AI) | Not measured | 39.4% | Only Hy4 measured |
| Vibe Code Bench (Vals AI) | Not measured | 81.0% | Only Hy4 measured |
| SciCode | 48.6% | Not measured | Only Hy3 measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 0.5% | 37.4% | Hy4 |
| FrontierCode 2.0 | Not measured | 31.4% | Only Hy4 measured |
| ALE-Bench | 838.6 | Not measured | Only Hy3 measured |
| LMArena Text: coding | 1467 | Not measured | Only Hy3 measured |
| LMArena Text: software and IT | 1464 | Not measured | Only Hy3 measured |
| LMArena WebDev: React | 1506 | Not measured | Only Hy3 measured |
| LMArena WebDev: HTML | 1488 | Not measured | Only Hy3 measured |
| GameBench 2 | Not measured | 48.7 | Only Hy4 measured |
| Terminal-Bench 4.0 | Not measured | 39.1% | Only Hy4 measured |
| Neither model has a result on the other 39 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| AA Intelligence Index | Not measured | 34.6 | Only Hy4 measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| Tax Agent Bench (Vals AI) | Not measured | 34.4% | Only Hy4 measured |
| Legal Research Bench (Vals AI) | Not measured | 41.4% | Only Hy4 measured |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 12.5% | Only Hy4 measured |
| Toolathlon-Verified | 56.8% | Not measured | Only Hy3 measured |
| Neither model has a result on the other 40 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Hy3 | Hy4 | 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 |
| HLE-Diamond (Artificial Analysis run) | 11.6% | 32.7% | Hy4 |
| CritPt v2, hard subset (Artificial Analysis run) | Not measured | 39.0% | Only Hy4 measured |
| LMArena Text: hard prompts | 1447 | Not measured | Only Hy3 measured |
| LMArena Text: expert | 1458 | Not measured | Only Hy3 measured |
| LMArena Text: hard prompts in English | 1453 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 19 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| LMArena Text: math | 1477 | Not measured | Only Hy3 measured |
| LMArena Text: mathematical work | 1480 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 12 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index5 boards, none shared
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| AA-Omniscience | Not measured | 5.4 | Only Hy4 measured |
| LMArena Text: science | 1455 | Not measured | Only Hy3 measured |
| LMArena Text: medicine and healthcare | 1449 | Not measured | Only Hy3 measured |
| LMArena Text: legal and government | 1425 | Not measured | Only Hy3 measured |
| LMArena Text: business and finance | 1425 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 16 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| Design Arena: websites | 1221 | 1251 | Hy4 |
| Design Arena: UI components | 1196 | 1197 | Hy4 |
| Design Arena: game dev | 1197 | 1271 | Hy4 |
| Design Arena: data viz | 1162 | Not measured | Only Hy3 measured |
| Design Arena: 3D | 1209 | 1251 | Hy4 |
| Design Arena: all code categories | 1215 | 1254 | Hy4 |
| Design Arena: SVG | 1191 | Not measured | Only Hy3 measured |
| Design Arena: ASCII art | 1155 | 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 index20 boards, none shared
| Benchmark | Hy3 | Hy4 | Leader |
|---|---|---|---|
| Vals Index | Not measured | 59.4% | Only Hy4 measured |
| Long context reasoning (AA-LCR) | 79.0% | Not measured | Only Hy3 measured |
| GDPval-AA | 1069 | Not measured | Only Hy3 measured |
| LMArena Text | 1439 | Not measured | Only Hy3 measured |
| LMArena Text: instruction following | 1426 | Not measured | Only Hy3 measured |
| LMArena Text: creative writing | 1405 | Not measured | Only Hy3 measured |
| LMArena Text: English | 1447 | Not measured | Only Hy3 measured |
| LMArena Text: non-English | 1426 | Not measured | Only Hy3 measured |
| LMArena Text: Chinese | 1490 | Not measured | Only Hy3 measured |
| LMArena Text: German | 1436 | Not measured | Only Hy3 measured |
| LMArena Text: Japanese | 1395 | Not measured | Only Hy3 measured |
| LMArena Text: Korean | 1403 | Not measured | Only Hy3 measured |
| LMArena Text: French | 1444 | Not measured | Only Hy3 measured |
| LMArena Text: Spanish | 1454 | Not measured | Only Hy3 measured |
| LMArena Text: Russian | 1438 | Not measured | Only Hy3 measured |
| LMArena Text: Polish | 1446 | Not measured | Only Hy3 measured |
| LMArena Text: multi-turn | 1438 | Not measured | Only Hy3 measured |
| LMArena Text: longer queries | 1443 | Not measured | Only Hy3 measured |
| LMArena Text: writing and language | 1416 | Not measured | Only Hy3 measured |
| LMArena Text: entertainment and media | 1405 | Not measured | Only Hy3 measured |
| Neither model has a result on the other 29 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Hy3 | Hy4 | 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 |
|---|---|---|
| Maker | Tencent | Tencent |
| Weights | Open weights | Open weights |
| Released | July 6, 2026 | November 4, 2026 |
| Context window | 262K | 1.05M |
| Input $/Mtok | $0.132 | $0.417 |
| Output $/Mtok | $0.528 | $1.2505 |
| Coding index | Not benchmarked | 65.2 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Hy3 better than Hy4 for coding?
Only Hy4 has a coding score, 65.2 on the Vibe Coding Index; Hy3 has none, so the two cannot be ranked on coding. Pick Hy3 for output tokens 2.4x cheaper ($0.528 vs $1.2505 per million).
Which is cheaper, Hy3 or Hy4?
Hy3 costs $0.132 per million input tokens and $0.528 per million output tokens; Hy4 costs $0.417 and $1.2505. Cost per solved task is known for neither of them.
Which has the bigger context window, Hy3 or Hy4?
Hy4, with 1.05M tokens against 262K.
How many benchmarks is the Hy3 vs Hy4 comparison based on?
7 benchmarks have a result for both models as of November 14, 2026. Hy3's coding index rests on 1 independent benchmark and Hy4's on 2. Where only one model has a result, the page says so instead of naming a winner.