Model comparison
Inkling vs Trinity Large Thinking
Inkling is the stronger model overall, scoring 44.1 to 38.6 on the Noometry Index. Trinity Large Thinking costs 6.6× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Last verified . 20 shared benchmarks.
Arcee AISummary
- They share 20 benchmarks with published results for both. Inkling scores higher in 7 categories and Trinity Large Thinking in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Inkling leads 40.4 to 16.9.
- The biggest single-benchmark swing is SciCode: 47% for Inkling and 36.1% for Trinity Large Thinking.
- Trinity Large Thinking is cheaper at $0.25 / $0.80 per million input/output tokens, against $1.87 / $4.68 for Inkling.
- Trinity Large Thinking accepts more context: 262K tokens versus 66K.
Side by side
| Inkling | Trinity Large Thinking | |
|---|---|---|
| Provider | Thinking Machines Lab | Arcee AI |
| Noometry Index | 44.1 | 38.6 |
| Released | 2026-07-15 | 2026-04-01 |
| Weights | Open | Open |
| Context window | 66K | 262K |
| Max output | 66K | 80K |
| Input $ / M tokens | $1.87 | $0.25 |
| Output $ / M tokens | $4.68 | $0.80 |
| Results tracked | 41 | 24 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Inkling: 34.5 (#234), Trinity Large Thinking: 34.1 (#244)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena WebDev | 1413 | 1238 |
| SciCode | 47% | 36.1% |
| LMArena Coding | 1464 | 1381 |
| FrontierCode | 14% | — |
| FrontierSWE | 4.1% | — |
| WeirdML | 32.3% | — |
| ALE-Bench | 946 | — |
Agentic & Tool Use Not comparable
Inkling: 29.6 (#85), Trinity Large Thinking: —
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| APEX-Agents | 33.8% | — |
| τ²-bench Banking | 25% | — |
Reasoning Inkling leads
Inkling: 40.4 (#56), Trinity Large Thinking: 16.9 (#298)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| CritPt | 5.4% | 0.9% |
| LMArena Hard Prompts | 1451 | 1350 |
| ARC-AGI-2 | 36.5% | — |
| SimpleBench | 50% | — |
| NYT Connections (extended) | — | 16.5% |
| ARC-AGI-1 | 79.5% | — |
| Chess Puzzles | 21% | — |
| Thematic Generalization | — | 41.6% |
| DTBench | 87.5% | — |
| LMCA | 37.6% | — |
| Surface Evolver Bench | — | 15.6% |
| Epoch Capabilities Index | 148.54 | — |
Math Trinity Large Thinking leads
Inkling: 31.3 (#225), Trinity Large Thinking: 37.6 (#149)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Math | 1479 | 1366 |
| FrontierMath (Tiers 1-3) | 33.3% | — |
| FrontierMath Tier 4 | 4.9% | — |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| ProofBench | 0% | — |
Knowledge Inkling leads
Inkling: 55.1 (#49), Trinity Large Thinking: 40.9 (#113)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Expert | 1465 | 1360 |
| GPQA Diamond | 88.3% | — |
| SimpleQA Verified | 40.3% | — |
| Vectara Hallucination Rate | — | 6.9% |
Multilingual Inkling leads
Inkling: 54.0 (#52), Trinity Large Thinking: 46.2 (#160)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Non-English | 1434 | 1325 |
| LMArena Chinese | 1490 | 1373 |
| LMArena French | 1458 | 1374 |
| LMArena German | 1446 | 1356 |
| LMArena Japanese | 1429 | 1311 |
| LMArena Korean | 1404 | 1306 |
| LMArena Russian | 1429 | 1337 |
| LMArena Spanish | 1448 | 1357 |
Instruction Following Inkling leads
Inkling: 75.1 (#71), Trinity Large Thinking: 70.5 (#162)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Instruction Following | 1426 | 1334 |
Long Context Inkling leads
Inkling: 43.8 (#86), Trinity Large Thinking: 41.3 (#144)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Longer Query | 1434 | 1355 |
Writing & Preference Inkling leads
Inkling: 65.2 (#51), Trinity Large Thinking: 53.8 (#158)
| Benchmark | Inkling | Trinity Large Thinking |
|---|---|---|
| LMArena Text | 1441 | 1340 |
| LMArena Creative Writing | 1387 | 1320 |
| LMArena Multi-Turn | 1436 | 1342 |
| EQ-Bench Creative Writing | 1611 | — |
| EQ-Bench 4 | 1226 | — |
Frequently asked questions
Is Inkling better than Trinity Large Thinking?
Inkling is the stronger model overall, scoring 44.1 to 38.6 on the Noometry Index. Trinity Large Thinking costs 6.6× less per token, which makes it the better buy when Inkling's lead doesn't matter for your workload.
Which is cheaper, Inkling 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; Inkling lists at $1.87 and $4.68.
Is Inkling or Trinity Large Thinking better for coding?
They score almost the same on coding (34.5 vs 34.1); test both on your own repository before choosing.
Which has the bigger context window?
Trinity Large Thinking does, with 262K tokens against 66K.
How many benchmarks do Inkling and Trinity Large Thinking share?
20 benchmarks have published results for both models. Inkling has 41 scored results on Noometry and Trinity Large Thinking has 24.