# Claude Sonnet 4.6 vs Trinity Large Thinking

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 38.1 on the Noometry Index. Trinity Large Thinking costs 15× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-6-vs-trinity-large-thinking
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 8 categories and Trinity Large Thinking in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 16.3.
- The biggest single-benchmark swing is NYT Connections (extended): 80.9% for Claude Sonnet 4.6 and 16.5% for Trinity Large Thinking.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 262K.
- Trinity Large Thinking has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| Provider | Anthropic | Arcee AI |
| Noometry Index | 50.1 | 38.1 |
| Rank | 42 | 180 |
| Context | 1M | 262K |
| Input $/M | $3 | $0.25 |
| Output $/M | $15 | $0.80 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 4.6: 46.0 (#62)
- Trinity Large Thinking: 33.8 (#238)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | 1521 | 1238 |
| SciCode | 46.8% | 36.1% |
| LMArena Coding | 1504 | 1381 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| WeirdML | 66.1% | — |
| ALE-Bench | 1,327 | — |

## Agentic & Tool Use

- Claude Sonnet 4.6: 39.1 (#26)
- Trinity Large Thinking: —

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| Vending-Bench 2 | 7,204 | — |

## Reasoning

- Claude Sonnet 4.6: 46.1 (#40)
- Trinity Large Thinking: 16.3 (#298)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| NYT Connections (extended) | 80.9% | 16.5% |
| CritPt | 3.1% | 0.9% |
| Thematic Generalization | 76.3% | 41.6% |
| LMArena Hard Prompts | 1485 | 1350 |
| ARC-AGI-2 | 60.4% | — |
| ARC-AGI-1 | 86.5% | — |
| Chess Puzzles | 13% | — |
| Mystery Game Puzzles | 16% | — |
| DTBench | 89.9% | — |
| LMCA | 46.5% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 152.25 | — |
| ForecastBench | 62 | — |

## Math

- Claude Sonnet 4.6: 51.9 (#41)
- Trinity Large Thinking: 34.9 (#149)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1463 | 1366 |
| OTIS Mock AIME 2024-2025 | 85.8% | — |
| ProofBench | 45% | — |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |

## Knowledge

- Claude Sonnet 4.6: 51.6 (#60)
- Trinity Large Thinking: 40.3 (#107)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| Vectara Hallucination Rate | 10.6% | 6.9% |
| LMArena Expert | 1500 | 1359 |
| GPQA Diamond | 87.4% | — |
| SimpleQA Verified | 35.5% | — |

## Multimodal

- Claude Sonnet 4.6: 38.0 (#61)
- Trinity Large Thinking: —

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |

## Multilingual

- Claude Sonnet 4.6: 54.6 (#39)
- Trinity Large Thinking: 46.4 (#155)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1440 | 1325 |
| LMArena Chinese | 1491 | 1372 |
| LMArena French | 1465 | 1374 |
| LMArena German | 1428 | 1356 |
| LMArena Japanese | 1419 | 1312 |
| LMArena Korean | 1411 | 1304 |
| LMArena Russian | 1440 | 1338 |
| LMArena Spanish | 1465 | 1356 |

## Instruction Following

- Claude Sonnet 4.6: 77.6 (#23)
- Trinity Large Thinking: 70.7 (#157)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1476 | 1334 |

## Long Context

- Claude Sonnet 4.6: 45.3 (#40)
- Trinity Large Thinking: 41.2 (#139)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1479 | 1355 |

## Writing & Preference

- Claude Sonnet 4.6: 70.4 (#22)
- Trinity Large Thinking: 54.0 (#152)

| Benchmark | Claude Sonnet 4.6 | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1458 | 1340 |
| LMArena Creative Writing | 1435 | 1321 |
| LMArena Multi-Turn | 1464 | 1342 |
| EQ-Bench Creative Writing | 1810 | — |
| EQ-Bench 4 | 1207 | — |

## FAQ

### Is Claude Sonnet 4.6 better than Trinity Large Thinking?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.1 to 38.1 on the Noometry Index. Trinity Large Thinking costs 15× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

### Which is cheaper, Claude Sonnet 4.6 or Trinity Large Thinking?

Trinity Large Thinking is cheaper. It lists at $0.25 per million input tokens and $0.80 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Claude Sonnet 4.6 or Trinity Large Thinking better for coding?

Claude Sonnet 4.6 scores higher on coding benchmarks: 46.0 versus 33.8 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 4.6 does, with 1M tokens against 262K.

### How many benchmarks do Claude Sonnet 4.6 and Trinity Large Thinking share?

23 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and Trinity Large Thinking has 24.