Trinity Large Thinking vs Hy3
Updated October 21, 2026. 43 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick Hy3 for coding: it scores 43.9 on the Vibe Coding Index against 5.4 for Trinity Large Thinking.
Trinity Large Thinking is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
- Shared results
- 43
- Trinity Large Thinking only
- 9
- Hy3 only
- 1
- Like-for-like categories
- 6 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
Hy3
Hy3 scores 43.9 against 5.4.
One benchmark behind each score
Intelligence
AA Intelligence Index
No call
Only Trinity Large Thinking has a AA Intelligence Index result, 10.8.
Directional only
Agentic work
AA Agentic Index
No call
Only Trinity Large Thinking has a AA Agentic Index result, 1.0.
Directional only
Web design
Design Arena: websites
Hy3
Hy3 leads, 1230 against 1137.
Same board, both measured
Long documents
Listed context window
No call
Both take 262K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Hy3
Trinity Large Thinking $0.0006, Hy3 $0.0004 per turn.
List prices
At a glance
- Coding index
- Trinity Large Thinking 5.4 from 1 benchmark, Hy3 43.9 from 1 benchmark.
- Widest category
- Coding, 28.2 points apart on average over 6 shared benchmarks.
- Widest single gap
- LMArena WebDev: React: Trinity Large Thinking 1231 against Hy3 1511, 46.7 points apart on a 0 to 100 scale.
- Output price
- Trinity Large Thinking $0.80 against Hy3 $0.528 per million output tokens, 1.5x apart.
- Context window
- Both take 262K tokens.
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 | Trinity Large Thinking | Bars | Hy3 | Ties |
|---|---|---|---|---|
| Coding | 0 | 6 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 4 | ||
| Web design | 0 | 8 | ||
| General | 0 | 18 | ||
| All shared boards | 0 | 43 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index06
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| LMArena WebDev | 1238 | 1507 | Hy3 |
| AA Coding Index | 25.8 | Not measured | Only Trinity Large Thinking measured |
| SciCode | 40.6% | 48.6% | Hy3 |
| Terminal-Bench Hard (Artificial Analysis run) | 22.7% | Not measured | Only Trinity Large Thinking measured |
| ALE-Bench | Not measured | 838.6 | Only Hy3 measured |
| LMArena Text: coding | 1357 | 1466 | Hy3 |
| LMArena Text: software and IT | 1362 | 1463 | Hy3 |
| LMArena WebDev: React | 1231 | 1511 | Hy3 |
| LMArena WebDev: HTML | 1240 | 1487 | Hy3 |
| Neither model has a result on the other 29 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 10.8 | Not measured | Only Trinity Large Thinking measured |
| Neither model has a result on the other 1 board here | |||
Agentic workDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| AA Agentic Index | 1.0 | Not measured | Only Trinity Large Thinking measured |
| τ²-Bench Telecom | 90.1% | Not measured | Only Trinity Large Thinking measured |
| Neither model has a result on the other 37 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | 75.2% | 89.7% | Hy3 |
| Humanity's Last Exam (Artificial Analysis run) | 15.8% | 33.5% | Hy3 |
| LMArena Text: hard prompts | 1342 | 1447 | Hy3 |
| LMArena Text: expert | 1359 | 1461 | Hy3 |
| LMArena Text: hard prompts in English | 1356 | 1455 | Hy3 |
| Neither model has a result on the other 16 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| LMArena Text: math | 1365 | 1473 | Hy3 |
| LMArena Text: mathematical work | 1350 | 1482 | Hy3 |
| Neither model has a result on the other 11 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index04
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| CaseLaw (Vals AI) | 57.9% | Not measured | Only Trinity Large Thinking measured |
| CorpFin (Vals AI) | 54.7% | Not measured | Only Trinity Large Thinking measured |
| LMArena Text: science | 1355 | 1457 | Hy3 |
| LMArena Text: medicine and healthcare | 1349 | 1452 | Hy3 |
| LMArena Text: legal and government | 1345 | 1422 | Hy3 |
| LMArena Text: business and finance | 1329 | 1424 | Hy3 |
| Neither model has a result on the other 11 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1137 | 1230 | Hy3 |
| Design Arena: UI components | 1045 | 1205 | Hy3 |
| Design Arena: game dev | 1079 | 1208 | Hy3 |
| Design Arena: data viz | 1098 | 1163 | Hy3 |
| Design Arena: 3D | 1082 | 1233 | Hy3 |
| Design Arena: all code categories | 1117 | 1227 | Hy3 |
| Design Arena: SVG | 1016 | 1199 | Hy3 |
| Design Arena: ASCII art | 1041 | 1157 | Hy3 |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index018
| Benchmark | Trinity Large Thinking | Hy3 | Leader |
|---|---|---|---|
| IFBench | 56.3% | Not measured | Only Trinity Large Thinking measured |
| Long context reasoning (AA-LCR) | 38.0% | 79.0% | Hy3 |
| Surface Evolver Bench | 15.6% | Not measured | Only Trinity Large Thinking measured |
| LMArena Text | 1340 | 1438 | Hy3 |
| LMArena Text: instruction following | 1326 | 1426 | Hy3 |
| LMArena Text: creative writing | 1308 | 1403 | Hy3 |
| LMArena Text: English | 1356 | 1448 | Hy3 |
| LMArena Text: non-English | 1325 | 1426 | Hy3 |
| LMArena Text: Chinese | 1371 | 1485 | Hy3 |
| LMArena Text: German | 1356 | 1449 | Hy3 |
| LMArena Text: Japanese | 1312 | 1397 | Hy3 |
| LMArena Text: Korean | 1306 | 1396 | Hy3 |
| LMArena Text: French | 1366 | 1456 | Hy3 |
| LMArena Text: Spanish | 1327 | 1457 | Hy3 |
| LMArena Text: Russian | 1338 | 1432 | Hy3 |
| LMArena Text: Polish | 1312 | 1447 | Hy3 |
| LMArena Text: multi-turn | 1320 | 1436 | Hy3 |
| LMArena Text: longer queries | 1318 | 1442 | Hy3 |
| LMArena Text: writing and language | 1317 | 1414 | Hy3 |
| LMArena Text: entertainment and media | 1296 | 1400 | Hy3 |
| Neither model has a result on the other 26 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Trinity Large Thinking | Hy3 | 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 | Arcee AI Trinity Large Thinking | Tencent Hy3 |
|---|---|---|
| Maker | Arcee AI | Tencent |
| Weights | Open weights | Open weights |
| Released | April 1, 2026 | July 6, 2026 |
| Context window | 262K | 262K |
| Input $/Mtok | $0.25 | $0.132 |
| Output $/Mtok | $0.80 | $0.528 |
| Coding index | 5.4 | 43.9 |
| Independent measurements behind it | 1 | 1 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Trinity Large Thinking better than Hy3 for coding?
Pick Hy3 for coding: it scores 43.9 on the Vibe Coding Index against 5.4 for Trinity Large Thinking. Trinity Large Thinking is neither cheaper nor larger in context, and leads no benchmark both models were measured on.
Which is cheaper, Trinity Large Thinking or Hy3?
Trinity Large Thinking costs $0.25 per million input tokens and $0.80 per million output tokens; Hy3 costs $0.132 and $0.528. Cost per solved task is known for neither of them.
Which has the bigger context window, Trinity Large Thinking or Hy3?
Neither: both take 262K tokens.
How many benchmarks is the Trinity Large Thinking vs Hy3 comparison based on?
43 benchmarks have a result for both models as of October 21, 2026. Trinity Large Thinking's coding index rests on 1 independent benchmark and Hy3's on 1. Where only one model has a result, the page says so instead of naming a winner.