Trinity Large Thinking vs Claude Sonnet 4.6
Claude Sonnet 4.6 leads Trinity Large Thinking on the Artificial Analysis Intelligence Index, 30.1 to 10.8, and Trinity Large Thinking costs 18.8x less per output token ($0.80 vs $15.00 per million). Updated October 21, 2026; 51 benchmarks have a result for both.
Trinity Large Thinking or Claude Sonnet 4.6: which to pick
Only Claude Sonnet 4.6 has a coding score, 57.3 on the Vibe Coding Index; Trinity Large Thinking has none, so the two cannot be ranked on coding.
Pick Trinity Large Thinking for output tokens 18.8x cheaper ($0.80 vs $15.00 per million) and work measured by τ²-Bench Telecom, where it leads.
- Shared results
- 51
- Trinity Large Thinking only
- 1
- Claude Sonnet 4.6 only
- 67
- Like-for-like categories
- 8 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 Claude Sonnet 4.6 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
Claude Sonnet 4.6
Claude Sonnet 4.6 leads, 30.1 against 10.8.
Same board, both measured
Agentic work
AA Agentic Index
Claude Sonnet 4.6
Claude Sonnet 4.6 leads, 31.8 against 1.0.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only Claude Sonnet 4.6 has a GPQA Diamond result, 85.6%.
Directional only
Math
AIME
No call
Only Claude Sonnet 4.6 has a AIME result, 92.3%.
Directional only
Knowledge
MMLU Pro
No call
Only Claude Sonnet 4.6 has a MMLU Pro result, 87.3%.
Directional only
Web design
Design Arena: websites
Claude Sonnet 4.6
Claude Sonnet 4.6 leads, 1285 against 1137.
Same board, both measured
Long documents
Listed context window
Claude Sonnet 4.6
Claude Sonnet 4.6 takes 1M tokens against 262K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Trinity Large Thinking
Trinity Large Thinking $0.0006, Claude Sonnet 4.6 $0.011 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
No call
Only Claude Sonnet 4.6 has the number, $1.68.
Directional only
At a glance
- Coding index
- Trinity Large Thinking not benchmarked, Claude Sonnet 4.6 57.3 from 5 benchmarks.
- Widest category
- Coding, 32.3 points apart on average over 8 shared benchmarks.
- Widest single gap
- LMArena WebDev: HTML: Trinity Large Thinking 1240 against Claude Sonnet 4.6 1527, 47.8 points apart on a 0 to 100 scale.
- Output price
- Trinity Large Thinking $0.80 against Claude Sonnet 4.6 $15.00 per million output tokens, 18.8x apart.
- Context window
- Trinity Large Thinking takes 262K tokens, Claude Sonnet 4.6 1M.
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 | Claude Sonnet 4.6 | Ties |
|---|---|---|---|---|
| Coding | 0 | 8 | ||
| Intelligence | 0 | 1 | ||
| Agentic work | 1 | 1 | ||
| Reasoning | 0 | 5 | ||
| Math | 0 | 2 | ||
| Knowledge | 0 | 6 | ||
| Web design | 0 | 8 | ||
| General | 0 | 19 | ||
| All shared boards | 1 | 50 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index08
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 66.6% | Only Claude Sonnet 4.6 measured |
| LMArena WebDev | 1238 | 1523 | Claude Sonnet 4.6 |
| AA Coding Index | 25.8 | 63.0 | Claude Sonnet 4.6 |
| LiveCodeBench (Vals AI) | Not measured | 82.1% | Only Claude Sonnet 4.6 measured |
| SWE-bench Verified (Vals AI) | Not measured | 77.4% | Only Claude Sonnet 4.6 measured |
| Vibe Code Bench (Vals AI) | Not measured | 55.8% | Only Claude Sonnet 4.6 measured |
| Code migration (Vals AI) | Not measured | 39.9% | Only Claude Sonnet 4.6 measured |
| ProgramBench (Vals AI) | Not measured | 0.5% | Only Claude Sonnet 4.6 measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 57.3% | Only Claude Sonnet 4.6 measured |
| SkillsBench (Vals AI) | Not measured | 49.0% | Only Claude Sonnet 4.6 measured |
| SciCode | 40.6% | 50.1% | Claude Sonnet 4.6 |
| Terminal-Bench Hard (Artificial Analysis run) | 22.7% | 53.0% | Claude Sonnet 4.6 |
| Terminal-Bench | Not measured | 53.4% | Only Claude Sonnet 4.6 measured |
| FrontierCode | Not measured | 24.3% | Only Claude Sonnet 4.6 measured |
| WeirdML | Not measured | 66.1% | Only Claude Sonnet 4.6 measured |
| LMArena Text: coding | 1357 | 1503 | Claude Sonnet 4.6 |
| LMArena Text: software and IT | 1362 | 1491 | Claude Sonnet 4.6 |
| LMArena WebDev: React | 1231 | 1512 | Claude Sonnet 4.6 |
| LMArena WebDev: HTML | 1240 | 1527 | Claude Sonnet 4.6 |
| LMArena WebDev: from an image | Not measured | 1529 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 19 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Intelligence Index | 10.8 | 30.1 | Claude Sonnet 4.6 |
| Epoch Capabilities Index | Not measured | 152.8 | Only Claude Sonnet 4.6 measured |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index11
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AA Agentic Index | 1.0 | 31.8 | Claude Sonnet 4.6 |
| Design Arena: agents building web apps | Not measured | 1190 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building full-stack apps | Not measured | 1213 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building mobile apps | Not measured | 1212 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Android apps | Not measured | 1196 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building Godot games | Not measured | 1226 | Only Claude Sonnet 4.6 measured |
| Design Arena: agents building games | Not measured | 1177 | Only Claude Sonnet 4.6 measured |
| LiveBench research | Not measured | 63.6% | Only Claude Sonnet 4.6 measured |
| LiveBench browser | Not measured | 68.0% | Only Claude Sonnet 4.6 measured |
| LiveBench long agentic | Not measured | 23.1% | Only Claude Sonnet 4.6 measured |
| Finance Agent (Vals AI) | Not measured | 51.0% | Only Claude Sonnet 4.6 measured |
| Tax Agent Bench (Vals AI) | Not measured | 25.8% | Only Claude Sonnet 4.6 measured |
| Legal Research Bench (Vals AI) | Not measured | 38.5% | Only Claude Sonnet 4.6 measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 60.2% | Only Claude Sonnet 4.6 measured |
| τ²-Bench Telecom | 90.1% | 79.5% | Trinity Large Thinking |
| APEX-Agents | Not measured | 43.0% | Only Claude Sonnet 4.6 measured |
| OSWorld 2.0 | Not measured | 9.3% | Only Claude Sonnet 4.6 measured |
| DeepResearch Bench | Not measured | 54.9% | Only Claude Sonnet 4.6 measured |
| OSWorld | Not measured | 72.1% | Only Claude Sonnet 4.6 measured |
| Vending-Bench 2 | Not measured | $7,204 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 21 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index05
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| GPQA Diamond | Not measured | 85.6% | Only Claude Sonnet 4.6 measured |
| LiveBench reasoning | Not measured | 86.4% | Only Claude Sonnet 4.6 measured |
| MMMU Pro (Vals AI) | Not measured | 83.6% | Only Claude Sonnet 4.6 measured |
| GPQA Diamond (Artificial Analysis run) | 75.2% | 87.5% | Claude Sonnet 4.6 |
| Humanity's Last Exam (Artificial Analysis run) | 15.8% | 33.6% | Claude Sonnet 4.6 |
| GPQA Diamond (Epoch AI run) | Not measured | 87.4% | Only Claude Sonnet 4.6 measured |
| ARC-AGI | Not measured | 86.5% | Only Claude Sonnet 4.6 measured |
| ARC-AGI-2 | Not measured | 60.4% | Only Claude Sonnet 4.6 measured |
| Chess puzzles | Not measured | 23.0% | Only Claude Sonnet 4.6 measured |
| Mystery game puzzles | Not measured | 16.0% | Only Claude Sonnet 4.6 measured |
| LMArena Text: hard prompts | 1342 | 1485 | Claude Sonnet 4.6 |
| LMArena Text: expert | 1359 | 1500 | Claude Sonnet 4.6 |
| LMArena Text: hard prompts in English | 1356 | 1493 | Claude Sonnet 4.6 |
| LMArena Vision: diagrams | Not measured | 1306 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: homework | Not measured | 1311 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 7 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| AIME | Not measured | 92.3% | Only Claude Sonnet 4.6 measured |
| OTIS Mock AIME | Not measured | 85.8% | Only Claude Sonnet 4.6 measured |
| ProofBench | Not measured | 45.0% | Only Claude Sonnet 4.6 measured |
| LMArena Text: math | 1365 | 1462 | Claude Sonnet 4.6 |
| LMArena Text: mathematical work | 1350 | 1483 | Claude Sonnet 4.6 |
| Neither model has a result on the other 8 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index06
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| MMLU Pro | Not measured | 87.3% | Only Claude Sonnet 4.6 measured |
| LegalBench (Vals AI) | Not measured | 82.1% | Only Claude Sonnet 4.6 measured |
| CaseLaw (Vals AI) | 57.9% | 64.0% | Claude Sonnet 4.6 |
| CorpFin (Vals AI) | 54.7% | 65.3% | Claude Sonnet 4.6 |
| TaxEval (Vals AI) | Not measured | 77.1% | Only Claude Sonnet 4.6 measured |
| MortgageTax (Vals AI) | Not measured | 67.7% | Only Claude Sonnet 4.6 measured |
| MedQA (Vals AI) | Not measured | 92.1% | Only Claude Sonnet 4.6 measured |
| SimpleQA Verified | Not measured | 35.5% | Only Claude Sonnet 4.6 measured |
| LMArena Text: science | 1355 | 1474 | Claude Sonnet 4.6 |
| LMArena Text: medicine and healthcare | 1349 | 1451 | Claude Sonnet 4.6 |
| LMArena Text: legal and government | 1345 | 1470 | Claude Sonnet 4.6 |
| LMArena Text: business and finance | 1329 | 1464 | Claude Sonnet 4.6 |
| FACTS Benchmark Suite | Not measured | 37.0% | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 5 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| Design Arena: websites | 1137 | 1285 | Claude Sonnet 4.6 |
| Design Arena: UI components | 1045 | 1272 | Claude Sonnet 4.6 |
| Design Arena: game dev | 1079 | 1259 | Claude Sonnet 4.6 |
| Design Arena: data viz | 1098 | 1277 | Claude Sonnet 4.6 |
| Design Arena: 3D | 1082 | 1234 | Claude Sonnet 4.6 |
| Design Arena: all code categories | 1117 | 1276 | Claude Sonnet 4.6 |
| Design Arena: SVG | 1017 | 1197 | Claude Sonnet 4.6 |
| Design Arena: ASCII art | 1041 | 1235 | Claude Sonnet 4.6 |
| LiveBench design | Not measured | 71.8% | Only Claude Sonnet 4.6 measured |
GeneralLike-for-likePublic boards, shown as published, kept out of the index019
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| LiveBench data analysis | Not measured | 76.1% | Only Claude Sonnet 4.6 measured |
| Public Benefits Bench (Vals AI) | Not measured | 62.5% | Only Claude Sonnet 4.6 measured |
| SAGE (Vals AI) | Not measured | 46.6% | Only Claude Sonnet 4.6 measured |
| Vals Multimodal Index | Not measured | 60.6% | Only Claude Sonnet 4.6 measured |
| IFBench | 56.3% | 56.6% | Claude Sonnet 4.6 |
| Long context reasoning (AA-LCR) | 38.0% | 80.0% | Claude Sonnet 4.6 |
| DTBench | Not measured | 89.9% | Only Claude Sonnet 4.6 measured |
| LMCA | Not measured | 46.5% | Only Claude Sonnet 4.6 measured |
| Surface Evolver Bench | 15.6% | Not measured | Only Trinity Large Thinking measured |
| ExploitBench | Not measured | 23.6% | Only Claude Sonnet 4.6 measured |
| LMArena Text | 1340 | 1458 | Claude Sonnet 4.6 |
| LMArena Text: instruction following | 1326 | 1476 | Claude Sonnet 4.6 |
| LMArena Text: creative writing | 1308 | 1435 | Claude Sonnet 4.6 |
| LMArena Vision | Not measured | 1283 | Only Claude Sonnet 4.6 measured |
| LMArena Text: English | 1356 | 1472 | Claude Sonnet 4.6 |
| LMArena Text: non-English | 1325 | 1440 | Claude Sonnet 4.6 |
| LMArena Text: Chinese | 1371 | 1491 | Claude Sonnet 4.6 |
| LMArena Text: German | 1356 | 1428 | Claude Sonnet 4.6 |
| LMArena Text: Japanese | 1312 | 1420 | Claude Sonnet 4.6 |
| LMArena Text: Korean | 1306 | 1410 | Claude Sonnet 4.6 |
| LMArena Text: French | 1366 | 1465 | Claude Sonnet 4.6 |
| LMArena Text: Spanish | 1327 | 1464 | Claude Sonnet 4.6 |
| LMArena Text: Russian | 1338 | 1440 | Claude Sonnet 4.6 |
| LMArena Text: Polish | 1312 | 1443 | Claude Sonnet 4.6 |
| LMArena Text: multi-turn | 1320 | 1464 | Claude Sonnet 4.6 |
| LMArena Text: longer queries | 1318 | 1479 | Claude Sonnet 4.6 |
| LMArena Text: writing and language | 1317 | 1444 | Claude Sonnet 4.6 |
| LMArena Text: entertainment and media | 1296 | 1431 | Claude Sonnet 4.6 |
| LMArena Vision: English | Not measured | 1298 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: Chinese | Not measured | 1307 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: reading text | Not measured | 1296 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: humor | Not measured | 1248 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: recognizing things | Not measured | 1195 | Only Claude Sonnet 4.6 measured |
| LMArena Vision: creative writing | Not measured | 1272 | Only Claude Sonnet 4.6 measured |
| LMArena Document | Not measured | 1482 | Only Claude Sonnet 4.6 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Trinity Large Thinking | Claude Sonnet 4.6 | 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
Trinity Large Thinking costs less
Repository review
50K input + 3K output tokens
Trinity Large Thinking costs less
Agent loop
200K input + 10K output tokens
Trinity Large Thinking 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 | Anthropic Claude Sonnet 4.6 |
|---|---|---|
| Maker | Arcee AI | Anthropic |
| Weights | Open weights | No public weights listed |
| Released | April 1, 2026 | February 17, 2026 |
| Context window | 262K | 1M |
| Input $/Mtok | $0.25 | $3.00 |
| Output $/Mtok | $0.80 | $15.00 |
| Coding index | Not benchmarked | 57.3 |
| Independent measurements behind it | 1 | 5 |
| Cost per solved task | Not computable | $1.68 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Trinity Large Thinking better than Claude Sonnet 4.6 for coding?
Only Claude Sonnet 4.6 has a coding score, 57.3 on the Vibe Coding Index; Trinity Large Thinking has none, so the two cannot be ranked on coding. Pick Trinity Large Thinking for output tokens 18.8x cheaper ($0.80 vs $15.00 per million) and work measured by τ²-Bench Telecom, where it leads.
Which is cheaper, Trinity Large Thinking or Claude Sonnet 4.6?
Trinity Large Thinking costs $0.25 per million input tokens and $0.80 per million output tokens; Claude Sonnet 4.6 costs $3.00 and $15.00. Cost per solved task is known for Claude Sonnet 4.6 only.
Which has the bigger context window, Trinity Large Thinking or Claude Sonnet 4.6?
Claude Sonnet 4.6, with 1M tokens against 262K.
How many benchmarks is the Trinity Large Thinking vs Claude Sonnet 4.6 comparison based on?
51 benchmarks have a result for both models as of October 21, 2026. Trinity Large Thinking's coding index rests on 1 independent benchmark and Claude Sonnet 4.6's on 5. Where only one model has a result, the page says so instead of naming a winner.